{"id":"book_method_c01_s01_p01","kind":"method","chapter":1,"chapter_title":"Introduction to linguistics problems","section":1,"section_title":"Structure of linguistics problems","topic":"general problem-solving methodology","text":"All linguistics problems have the same structure, consisting of four parts (excluding the title and the author):\n\n- Introduction\n\nIn the introduction, we learn about the language featured in the problem. For most problems, the introduction is simply a sentence like ``Given below are some [words/structures/constructions/sentences, etc.] in [Language] and their English translations (in random order).\"\n\nSome problems might have a more complex introduction which also includes information about the culture of the people speaking that language or information about certain characteristics of the language. Generally speaking, if the introduction is complex and contains additional information, this information is likely to be relevant to solving the problem.\n\n- Dataset\n\nThis part contains the examples based on which we should solve the tasks. This part is also known as the corpus. If the corpus and the translations are given in order (it is known which translation corresponds to which structure), the problem is called a Rosetta stone problem, while if the translations are given in random order, we are talking about a chaos-and-order problem.\n\n- Tasks\n\nThe corpus is followed by the tasks. For chaos-and-order problems, the first task will always be “Determine the correct correspondences”. Therefore, an easy way to figure out whether the problem is Rosetta stone or chaos-and-order (besides reading the introduction) is by checking the first task.\n\nThe subsequent tasks will be “Translate into English” and “Translate into […]”. As a rule of thumb, the task asking to translate into English will precede the other one because (1) in order to translate into English, it is not always necessary to understand all the grammar rules, and (2) we might be able to use these additional examples in order to gather more information.\n\nSome problems might also have special tasks, which usually offer important hints about the phenomena featured in the problem.\n\n- Notes\n\nAt the end of each problem, there will be some notes which provide three types of information. Firstly, there will be some data about the language featured in the problem, such as where it is spoken, how many people it is spoken by, what language family it belongs to, etc. In general, this information is not relevant to solving the problem, although, for experienced solvers, the family to which the language belongs might offer additional helpful information. In this book, this information has been removed from the notes and all the information regarding the languages can be found in Appendices appendix:1 and appendix:2.\n\nRelevant phonetic information follows, which offers details regarding how some letters, characters, symbols are pronounced. Moreover, it might also include details regarding the existence of diphthongs, or details on additional writing notations used in the problem. Depending on the problem, this kind of information may or may not be useful.\n\nFinally, there might be some information about specific words, usually referring to types or species of animals or plants, traditional objects or garments, etc. Again, depending on the type of problem, this information may or may not be useful in solving it.","source_file":"chapters/01-IntroToLO.tex","source_line_start":7,"license":"CC-BY-4.0","retrieval_text":"Introduction to linguistics problems\nStructure of linguistics problems\ngeneral problem-solving methodology\nAll linguistics problems have the same structure, consisting of four parts (excluding the title and the author):\n\n- Introduction\n\nIn the introduction, we learn about the language featured in the problem. For most problems, the introduction is simply a sentence like ``Given below are some [words/structures/constructions/sentences, etc.] in [Language] and their English translations (in random order).\"\n\nSome problems might have a more complex introduction which also includes information about the culture of the people speaking that language or information about certain characteristics of the language. Generally speaking, if the introduction is complex and contains additional information, this information is likely to be relevant to solving the problem.\n\n- Dataset\n\nThis part contains the examples based on which we should solve the tasks. This part is also known as the corpus. If the corpus and the translations are given in order (it is known which translation corresponds to which structure), the problem is called a Rosetta stone problem, while if the translations are given in random order, we are talking about a chaos-and-order problem.\n\n- Tasks\n\nThe corpus is followed by the tasks. For chaos-and-order problems, the first task will always be “Determine the correct correspondences”. Therefore, an easy way to figure out whether the problem is Rosetta stone or chaos-and-order (besides reading the introduction) is by checking the first task.\n\nThe subsequent tasks will be “Translate into English” and “Translate into […]”. As a rule of thumb, the task asking to translate into English will precede the other one because (1) in order to translate into English, it is not always necessary to understand all the grammar rules, and (2) we might be able to use these additional examples in order to gather more information.\n\nSome problems might also have special tasks, which usually offer important hints about the phenomena featured in the problem.\n\n- Notes\n\nAt the end of each problem, there will be some notes which provide three types of information. Firstly, there will be some data about the language featured in the problem, such as where it is spoken, how many people it is spoken by, what language family it belongs to, etc. In general, this information is not relevant to solving the problem, although, for experienced solvers, the family to which the language belongs might offer additional helpful information. In this book, this information has been removed from the notes and all the information regarding the languages can be found in Appendices appendix:1 and appendix:2.\n\nRelevant phonetic information follows, which offers details regarding how some letters, characters, symbols are pronounced. Moreover, it might also include details regarding the existence of diphthongs, or details on additional writing notations used in the problem. Depending on the problem, this kind of information may or may not be useful.\n\nFinally, there might be some information about specific words, usually referring to types or species of animals or plants, traditional objects or garments, etc. Again, depending on the type of problem, this information may or may not be useful in solving it."}
{"id":"book_method_c01_s02_p01","kind":"method","chapter":1,"chapter_title":"Introduction to linguistics problems","section":2,"section_title":"Classification of linguistics problems","topic":"general problem-solving methodology","text":"Generally, linguistics problems are classified based on the main phenomenon that is featured (which is closely related to a specific field of linguistics). In this book, the problems are divided into seven categories, as follows:\n\n- Writing systems (chap-writing)\n\nThese problems feature words in an unfamiliar writing system together with their transliterations in Latin script. These problems can be either Rosetta stone or chaos-and-order.\n\n- Phonetics and Phonology (Chapters chap-phonetics and chap-phonology)\n\nThese problems are based on the sound changes that occur in different environments or contexts (e.g., two or more forms of a word – such as singular–plural, different noun cases, declensions, etc. –, the way certain words change in different dialects, how words are transcribed phonetically, how words are stressed, etc.). Generally, these problems are Rosetta stone.\n\n- Morphology (Chapters chap-noun and chap-verb)\n\nThese problems are subdivided into two categories: morphology of the noun and its relation to other elements in the noun phrase (chap-noun) and morphology of the verb and its relation to other elements in the verb phrase (chap-verb). These problems can be either Rosetta stone or chaos-and-order.\n\n- Syntax (chap-syntax)\n\nSyntax problems contain sentences (or phrases), combining, to some extent, the morphology of the noun with the morphology of the verb. In most cases, these problems are Rosetta stone, mainly for simplicity. If the sentences were in random order, once the correspondences are made, all sentences need to be copied again – with their corresponding translations – in order to better see all the data and extract all the rules and phenomena that occur; the copying of the sentences would require a lot of time.\n\n- Semantics (chap-semantics)\n\nSemantic problems are not based on grammar rules, but rather on associations of words which share different properties related to their meaning. These problems are always chaos-and-order problems.\n\n- Number systems (chap-numbers)\n\nThis kind of problem contains numbers written out in a certain language and their numeric representation. The ``translations'' might or might not be ordered. Sometimes, the corpus might consist solely of some mathematical equalities, in which the numbers are written out in a certain language.\n\n- Other types of problems\n\nThis category is rather large and includes any problem that does not fit in the aforementioned categories. Nevertheless, even in this category, we can notice some types of problems that appear frequently in the linguistics olympiads. As a result, we can differentiate:\n\n- Metrics and prosody (chap-phonetics) – in which the corpus consists of lines from different poems and the purpose is discovering the general structure of the verse.\n\n- Time problems (chap-numbers) – these problems highlight how the calendar dates or the time are told in different languages.\n\n- Kinship problems (chap-others).\n\n- Orientation system problems (chap-others) – which show how directionality is represented in different languages.\n\nOf course, there can also be mixed problems, which combine two (or more) of the above categories.\n\nStatistically speaking, based on 437 problems from different national and international linguistics olympiads, we notice that the most common type of problem is the syntax one (accounting for approximately 19.5% of the problems). Indeed, syntax problems are usually pervasive in all olympiads and each olympiad will have at least one syntax problem. Morphology problems (both nominal and verbal) are almost as common as syntax ones, accounting for 19% of the total. The other categories are phonetics and phonology (13.5%), writing systems (12%), number systems (9%) and semantics (7%).","source_file":"chapters/01-IntroToLO.tex","source_line_start":34,"license":"CC-BY-4.0","retrieval_text":"Introduction to linguistics problems\nClassification of linguistics problems\ngeneral problem-solving methodology\nGenerally, linguistics problems are classified based on the main phenomenon that is featured (which is closely related to a specific field of linguistics). In this book, the problems are divided into seven categories, as follows:\n\n- Writing systems (chap-writing)\n\nThese problems feature words in an unfamiliar writing system together with their transliterations in Latin script. These problems can be either Rosetta stone or chaos-and-order.\n\n- Phonetics and Phonology (Chapters chap-phonetics and chap-phonology)\n\nThese problems are based on the sound changes that occur in different environments or contexts (e.g., two or more forms of a word – such as singular–plural, different noun cases, declensions, etc. –, the way certain words change in different dialects, how words are transcribed phonetically, how words are stressed, etc.). Generally, these problems are Rosetta stone.\n\n- Morphology (Chapters chap-noun and chap-verb)\n\nThese problems are subdivided into two categories: morphology of the noun and its relation to other elements in the noun phrase (chap-noun) and morphology of the verb and its relation to other elements in the verb phrase (chap-verb). These problems can be either Rosetta stone or chaos-and-order.\n\n- Syntax (chap-syntax)\n\nSyntax problems contain sentences (or phrases), combining, to some extent, the morphology of the noun with the morphology of the verb. In most cases, these problems are Rosetta stone, mainly for simplicity. If the sentences were in random order, once the correspondences are made, all sentences need to be copied again – with their corresponding translations – in order to better see all the data and extract all the rules and phenomena that occur; the copying of the sentences would require a lot of time.\n\n- Semantics (chap-semantics)\n\nSemantic problems are not based on grammar rules, but rather on associations of words which share different properties related to their meaning. These problems are always chaos-and-order problems.\n\n- Number systems (chap-numbers)\n\nThis kind of problem contains numbers written out in a certain language and their numeric representation. The ``translations'' might or might not be ordered. Sometimes, the corpus might consist solely of some mathematical equalities, in which the numbers are written out in a certain language.\n\n- Other types of problems\n\nThis category is rather large and includes any problem that does not fit in the aforementioned categories. Nevertheless, even in this category, we can notice some types of problems that appear frequently in the linguistics olympiads. As a result, we can differentiate:\n\n- Metrics and prosody (chap-phonetics) – in which the corpus consists of lines from different poems and the purpose is discovering the general structure of the verse.\n\n- Time problems (chap-numbers) – these problems highlight how the calendar dates or the time are told in different languages.\n\n- Kinship problems (chap-others).\n\n- Orientation system problems (chap-others) – which show how directionality is represented in different languages.\n\nOf course, there can also be mixed problems, which combine two (or more) of the above categories.\n\nStatistically speaking, based on 437 problems from different national and international linguistics olympiads, we notice that the most common type of problem is the syntax one (accounting for approximately 19.5% of the problems). Indeed, syntax problems are usually pervasive in all olympiads and each olympiad will have at least one syntax problem. Morphology problems (both nominal and verbal) are almost as common as syntax ones, accounting for 19% of the total. The other categories are phonetics and phonology (13.5%), writing systems (12%), number systems (9%) and semantics (7%)."}
{"id":"book_method_c01_s03_p01","kind":"method","chapter":1,"chapter_title":"Introduction to linguistics problems","section":3,"section_title":"Understanding the problem","topic":"general problem-solving methodology","text":"When solving a linguistics problem, it is important to understand what exactly is expected from us. In most cases, the way the tasks are phrased (especially the special tasks, as mentioned above), but even the way the corpus is chosen, can offer important hints. For example, let us take a look at the way the following task is phrased (the whole problem is presented in chap-noun, Problem 5.13):\n\n- Fill in the blanks:\n\n. | baqra [blank] | ‘blue cow’\n. | fjuri [blank] | ‘red flowers’\n. | kelb [blank] | ‘brown dog’\n. | kotba [blank] | ‘yellow books’\n. | siġra [blank] | ‘green tree’\n. | mwejjed [blank] | ‘purple chairs’\n. | tuffieħa [blank] | ‘yellow apple’\n\nThe thing that strikes us the most is that the first word is always given, and we only need to be concerned about the second. Normally, in a classical problem, we would simply be asked to translate the structures ‘blue cow’, ‘red flowers’, etc. Since in this case the first word (which, from the full dataset – not shown here –, we can see evidently represents the noun) is already given, we infer that, most likely, in this language (or at least based on the information given) the noun plural formation is irregular (or simply too complex) and does not follow specific rules, thus not being able to infer the singular form from the plural form or vice versa. Therefore, since the nouns are already mentioned, we know that the core phenomenon of the problem focuses on the adjective, rather than on the noun.\n\nLet us consider the following (fictitious) example which would correspond to a writing system problem:\n\n- Write in the [...] script: Mars, venus, JUPITER, NePtUne.\n\nIn this case, we deem unusual the writing of these words, some of them being just lowercase, others just uppercase, and others being written with a combination of the two. There is no plausible reason to do so unless the writing system differentiates between lower- and uppercase letters. Therefore, in this case, the choice of the tasks (and their form) offers us a valuable clue: most likely there is a difference between lower- and uppercase letters.\n\nAnother task might be:From a problem by Bozhidar Bozhanov (UKLO 2010).\n\n- Knowing that in Turkish dil = ‘language’, translate ‘linguist’, ‘mute’.\n\nIn this case, we are given a new word (‘language’) and we are asked to translate two words which belong to the same semantic field. Therefore, in order to translate ‘linguist’, there must be a rule which allows the formation of an agent noun (the person which...) and similarly, in the case of ‘mute’, we need to find a “negative”-forming particle (which marks the impossibilitylack ofincapacity etc.), both of which must be inferred from the data given.\n\nAs a result, it is important to read all the data and tasks carefully and ask ourselves whenever we see something slightly peculiar: Why is it like this?","source_file":"chapters/01-IntroToLO.tex","source_line_start":71,"license":"CC-BY-4.0","retrieval_text":"Introduction to linguistics problems\nUnderstanding the problem\ngeneral problem-solving methodology\nWhen solving a linguistics problem, it is important to understand what exactly is expected from us. In most cases, the way the tasks are phrased (especially the special tasks, as mentioned above), but even the way the corpus is chosen, can offer important hints. For example, let us take a look at the way the following task is phrased (the whole problem is presented in chap-noun, Problem 5.13):\n\n- Fill in the blanks:\n\n. | baqra [blank] | ‘blue cow’\n. | fjuri [blank] | ‘red flowers’\n. | kelb [blank] | ‘brown dog’\n. | kotba [blank] | ‘yellow books’\n. | siġra [blank] | ‘green tree’\n. | mwejjed [blank] | ‘purple chairs’\n. | tuffieħa [blank] | ‘yellow apple’\n\nThe thing that strikes us the most is that the first word is always given, and we only need to be concerned about the second. Normally, in a classical problem, we would simply be asked to translate the structures ‘blue cow’, ‘red flowers’, etc. Since in this case the first word (which, from the full dataset – not shown here –, we can see evidently represents the noun) is already given, we infer that, most likely, in this language (or at least based on the information given) the noun plural formation is irregular (or simply too complex) and does not follow specific rules, thus not being able to infer the singular form from the plural form or vice versa. Therefore, since the nouns are already mentioned, we know that the core phenomenon of the problem focuses on the adjective, rather than on the noun.\n\nLet us consider the following (fictitious) example which would correspond to a writing system problem:\n\n- Write in the [...] script: Mars, venus, JUPITER, NePtUne.\n\nIn this case, we deem unusual the writing of these words, some of them being just lowercase, others just uppercase, and others being written with a combination of the two. There is no plausible reason to do so unless the writing system differentiates between lower- and uppercase letters. Therefore, in this case, the choice of the tasks (and their form) offers us a valuable clue: most likely there is a difference between lower- and uppercase letters.\n\nAnother task might be:From a problem by Bozhidar Bozhanov (UKLO 2010).\n\n- Knowing that in Turkish dil = ‘language’, translate ‘linguist’, ‘mute’.\n\nIn this case, we are given a new word (‘language’) and we are asked to translate two words which belong to the same semantic field. Therefore, in order to translate ‘linguist’, there must be a rule which allows the formation of an agent noun (the person which...) and similarly, in the case of ‘mute’, we need to find a “negative”-forming particle (which marks the impossibilitylack ofincapacity etc.), both of which must be inferred from the data given.\n\nAs a result, it is important to read all the data and tasks carefully and ask ourselves whenever we see something slightly peculiar: Why is it like this?"}
{"id":"book_method_c01_s04_p01","kind":"method","chapter":1,"chapter_title":"Introduction to linguistics problems","section":4,"section_title":"Solution writing","topic":"general problem-solving methodology","text":"Solution writing is a core part of problem-solving. It is important to write all the rules clearly and consistently and to cover all the phenomena featured, but, at the same time, to write them succinctly enough to not waste precious time during an official competition.\n\nOfficially, in the guidelines of the IOL (and in the case of most linguistics competitions) it is stated that: “Unless stated differently, you should describe any patterns or rules that you identified in the data. Otherwise, your solution will not be awarded full marks.”\n\nThe most important thing we need to understand is that we need to write the rules that we identified and not how we found them. Therefore, in the linguistics competitions, we are not asked to provide our reasoning for inferring the rules, but rather only to write the actual rules.\n\nWhen writing the solution, we should:\n\n- use tables, graphs, diagrams or any other kind of concise representation;\n\n- use common abbreviations and symbols (we may also use less common abbreviations as long as we make a legend describing what they stand for);\n\n- explain all the rules and phenomena that occur.\n\nBriefly, through orderly and concise writing of all the rules, we ought to try and tell the story of the language we discovered.\n\nAt the same time, when writing the solution, we should not:\n\n- explain how we inferred or deduced the rules and patterns;\n\n- write dictionaries and explain the meaning of every single word (we will talk more about this in the next section);\n\n- use connectors and excessive words, such as: ``I think\", ``we deduce\", ``it is obvious that\", ``since\", ``therefore\", ``because\", ``it is possible that\", etc.\n\nTherefore, when providing a solution, we should not write something like “Since examples 1, 3, and 5 all contain the word mi and all the English translations of these examples end in a question mark (thus being interrogative constructions), we most likely can infer that this word marks the fact that the structure is an interrogative one” since the same explanation can (and should) be briefly written as: “mi = question”.","source_file":"chapters/01-IntroToLO.tex","source_line_start":109,"license":"CC-BY-4.0","retrieval_text":"Introduction to linguistics problems\nSolution writing\ngeneral problem-solving methodology\nSolution writing is a core part of problem-solving. It is important to write all the rules clearly and consistently and to cover all the phenomena featured, but, at the same time, to write them succinctly enough to not waste precious time during an official competition.\n\nOfficially, in the guidelines of the IOL (and in the case of most linguistics competitions) it is stated that: “Unless stated differently, you should describe any patterns or rules that you identified in the data. Otherwise, your solution will not be awarded full marks.”\n\nThe most important thing we need to understand is that we need to write the rules that we identified and not how we found them. Therefore, in the linguistics competitions, we are not asked to provide our reasoning for inferring the rules, but rather only to write the actual rules.\n\nWhen writing the solution, we should:\n\n- use tables, graphs, diagrams or any other kind of concise representation;\n\n- use common abbreviations and symbols (we may also use less common abbreviations as long as we make a legend describing what they stand for);\n\n- explain all the rules and phenomena that occur.\n\nBriefly, through orderly and concise writing of all the rules, we ought to try and tell the story of the language we discovered.\n\nAt the same time, when writing the solution, we should not:\n\n- explain how we inferred or deduced the rules and patterns;\n\n- write dictionaries and explain the meaning of every single word (we will talk more about this in the next section);\n\n- use connectors and excessive words, such as: ``I think\", ``we deduce\", ``it is obvious that\", ``since\", ``therefore\", ``because\", ``it is possible that\", etc.\n\nTherefore, when providing a solution, we should not write something like “Since examples 1, 3, and 5 all contain the word mi and all the English translations of these examples end in a question mark (thus being interrogative constructions), we most likely can infer that this word marks the fact that the structure is an interrogative one” since the same explanation can (and should) be briefly written as: “mi = question”."}
{"id":"book_method_c01_s05_p01","kind":"method","chapter":1,"chapter_title":"Introduction to linguistics problems","section":5,"section_title":"Dictionary vs. rules","topic":"general problem-solving methodology","text":"We mentioned above that a solution should not include the dictionary. By dictionary we mean the base words (or stems) such as nouns, pronouns, adverbs, etc., whose form does not change.\n\nOn the other hand, the rules (which we need to write) explain the alternations that occur in the language. Therefore, they explain the word order, the way words change depending on their number, gender, tense, etc. We also include here all the words that do not have a direct English translation (usually they represent words that have a function rather than a meaning). For example, in Chinese, the character 吗 placed at the end of the sentence signals that it is a direct question (requiring a yes/no answer). Therefore, since this character has a function (marks the interrogation) and not a meaning (it does not mean anything and it would not be found in a dictionary), it must be included in the solution writing.\n\nLet us consider the following dataset from the Turkish language, and imagine we are asked about how possession is marked in Turkish:\n\nbabam | ‘my father’\nkedin | ‘yoursg cat’\nkedimiz | ‘our cat’\nbaban | ‘yoursg father’\nkedi | ‘his cat’\n\nIn this case, we can easily observe that the possessive is marked with a suffix (attached at the end of the word) as follows: -m for ‘my’, -n for ‘yoursg’, -miz for ‘our’, while for ‘his’ nothing is added – in fact, it is important to specify that ``zero\" is added or, in other words, that a null morpheme is used to mark the equivalent of ‘his’ in English.\n\nFor this problem, the dictionary is: baba = ‘father’ and kedi = ‘cat’ (these words are invariable). Therefore, a correct and complete way of writing the solution is:\n\npossession suffixes: -m for ‘my’, -n for ‘yoursg’, -miz for ‘our’, for ‘his’ NotebulbonThe symbol marks the fact that nothing is added (it represents the null morpheme). This symbol does not need an explanation/legend when used.\n\nAnother, briefer, way to write the solution is:\n\nex:1:turkish\npossession suffixes: -m = 1sg, -n = 2sg, -miz = 1pl, = 3sg\n\nThe simplest way to write these suffixes is by creating a table which includes the persons (1, 2, 3) in the rows and the numbers in the columns (singular, plural). Therefore, we can also write:\n\nex:1:turkish\n\npossession suffixes:\n| sg | pl\n1 | -m | -miz\n2 | -n |\n3 | |\n\nIn this case, we can see the importance of using the null morpheme (). Thanks to it, we can deduce (based on the table above) that there is a difference between the possessive suffix for 2pl (the cell is blank – therefore we cannot deduce it based on the data given) and for 3sg (the cell is not empty, it contains the symbol , thus proving that we discovered the way it is marked).\n\nLet us now consider the following three sentences in Spanish, and imagine we are asked to describe the word order:\n\nTu marido corre. | = | ‘Yoursg husband runs.’\nÉl ve a tu marido. | = | ‘He sees yoursg husband.’\nMi novio ve a él. | = | ‘My boyfriend sees him.’\n\nThe rules we need to write concern the order of subject, verb and object and the order of the possessor and the possessed. Therefore, the solution is:\n\nWord order: S V (a O); Possessor – Possessed\nThe rules above contain a lot of relevant information, written very briefly:\n\n- The word order is S(ubject), followed by V(erb), followed by the particle a and finally followed by the O(bject);\n\n- The particle a only appears together with the object; if the sentence has no object, the particle a is not used, a fact marked by the use of brackets around the structure;\n\n- In a possessive construction, the possessor (owner) is placed before the possessed object.","source_file":"chapters/01-IntroToLO.tex","source_line_start":134,"license":"CC-BY-4.0","retrieval_text":"Introduction to linguistics problems\nDictionary vs. rules\ngeneral problem-solving methodology\nWe mentioned above that a solution should not include the dictionary. By dictionary we mean the base words (or stems) such as nouns, pronouns, adverbs, etc., whose form does not change.\n\nOn the other hand, the rules (which we need to write) explain the alternations that occur in the language. Therefore, they explain the word order, the way words change depending on their number, gender, tense, etc. We also include here all the words that do not have a direct English translation (usually they represent words that have a function rather than a meaning). For example, in Chinese, the character 吗 placed at the end of the sentence signals that it is a direct question (requiring a yes/no answer). Therefore, since this character has a function (marks the interrogation) and not a meaning (it does not mean anything and it would not be found in a dictionary), it must be included in the solution writing.\n\nLet us consider the following dataset from the Turkish language, and imagine we are asked about how possession is marked in Turkish:\n\nbabam | ‘my father’\nkedin | ‘yoursg cat’\nkedimiz | ‘our cat’\nbaban | ‘yoursg father’\nkedi | ‘his cat’\n\nIn this case, we can easily observe that the possessive is marked with a suffix (attached at the end of the word) as follows: -m for ‘my’, -n for ‘yoursg’, -miz for ‘our’, while for ‘his’ nothing is added – in fact, it is important to specify that ``zero\" is added or, in other words, that a null morpheme is used to mark the equivalent of ‘his’ in English.\n\nFor this problem, the dictionary is: baba = ‘father’ and kedi = ‘cat’ (these words are invariable). Therefore, a correct and complete way of writing the solution is:\n\npossession suffixes: -m for ‘my’, -n for ‘yoursg’, -miz for ‘our’, for ‘his’ NotebulbonThe symbol marks the fact that nothing is added (it represents the null morpheme). This symbol does not need an explanation/legend when used.\n\nAnother, briefer, way to write the solution is:\n\nex:1:turkish\npossession suffixes: -m = 1sg, -n = 2sg, -miz = 1pl, = 3sg\n\nThe simplest way to write these suffixes is by creating a table which includes the persons (1, 2, 3) in the rows and the numbers in the columns (singular, plural). Therefore, we can also write:\n\nex:1:turkish\n\npossession suffixes:\n| sg | pl\n1 | -m | -miz\n2 | -n |\n3 | |\n\nIn this case, we can see the importance of using the null morpheme (). Thanks to it, we can deduce (based on the table above) that there is a difference between the possessive suffix for 2pl (the cell is blank – therefore we cannot deduce it based on the data given) and for 3sg (the cell is not empty, it contains the symbol , thus proving that we discovered the way it is marked).\n\nLet us now consider the following three sentences in Spanish, and imagine we are asked to describe the word order:\n\nTu marido corre. | = | ‘Yoursg husband runs.’\nÉl ve a tu marido. | = | ‘He sees yoursg husband.’\nMi novio ve a él. | = | ‘My boyfriend sees him.’\n\nThe rules we need to write concern the order of subject, verb and object and the order of the possessor and the possessed. Therefore, the solution is:\n\nWord order: S V (a O); Possessor – Possessed\nThe rules above contain a lot of relevant information, written very briefly:\n\n- The word order is S(ubject), followed by V(erb), followed by the particle a and finally followed by the O(bject);\n\n- The particle a only appears together with the object; if the sentence has no object, the particle a is not used, a fact marked by the use of brackets around the structure;\n\n- In a possessive construction, the possessor (owner) is placed before the possessed object."}
{"id":"book_method_c01_s05_p02","kind":"method","chapter":1,"chapter_title":"Introduction to linguistics problems","section":5,"section_title":"Dictionary vs. rules","topic":"general problem-solving methodology","text":"- In a possessive construction, the possessor (owner) is placed before the possessed object.\n\nWe need not mention anything about the verbs since all of them are in the third person singular (3sg) present tense: therefore, we do not know how (or if) they change form in any way.\n\nLet us now consider three more sentences (as a supplement to those above):\n\nÉl ve tu casa. | = | ‘He sees yoursg house.’\nElla ve a su padre. | = | ‘She sees her father.’\nYo veo tu libro. | = | ‘I see yoursg book.’\n\nBased on these extra sentences, it is important to check, first of all, whether the rules we wrote before still hold for these examples as well. Therefore, we now see that a does not appear every time before the object, but it is only used when the object is human. Therefore, the rules become:\n\nex:1:spanish\nWord order: S V O; Possessor – Possessed\nIf O = person, add a before it.\n\nMoreover, we see this time that the verb changes, having the pair ve (‘he/she sees’) and veo (‘I see’). Therefore, we also need to pay attention to verbal morphology. If we do so, we can deduce the conjugation rules for the verb: -o = 1sg, = 3sg.\n\nThus, the final rules are:\n\nex:1:spanish\n\nWord order: S V O; Possessor – Possessed\n\nIf O = person, add a before it.\n\nVerb: 1sg = -o, 3sg =\n\nLet us now consider the following possible task:\n\n- There is an error in the following data. What is it?\n\nMi padre correo. | = | ‘My father runs.’\n\nSince we are told that there is an error in the sentence, we need to check the rules we have in order to see if any of them could justify this task. We remember that the -o ending of verbs is for 1sg subjects, but here the subject is 3sg, meaning there should be zero inflection on the verb, so the Spanish sentence should read Mi padre corre.\n\nIn the following chapters, each problem will be accompanied by a solution so that the reader can get accustomed to different ways of writing the solution.","source_file":"chapters/01-IntroToLO.tex","source_line_start":134,"license":"CC-BY-4.0","retrieval_text":"Introduction to linguistics problems\nDictionary vs. rules\ngeneral problem-solving methodology\n- In a possessive construction, the possessor (owner) is placed before the possessed object.\n\nWe need not mention anything about the verbs since all of them are in the third person singular (3sg) present tense: therefore, we do not know how (or if) they change form in any way.\n\nLet us now consider three more sentences (as a supplement to those above):\n\nÉl ve tu casa. | = | ‘He sees yoursg house.’\nElla ve a su padre. | = | ‘She sees her father.’\nYo veo tu libro. | = | ‘I see yoursg book.’\n\nBased on these extra sentences, it is important to check, first of all, whether the rules we wrote before still hold for these examples as well. Therefore, we now see that a does not appear every time before the object, but it is only used when the object is human. Therefore, the rules become:\n\nex:1:spanish\nWord order: S V O; Possessor – Possessed\nIf O = person, add a before it.\n\nMoreover, we see this time that the verb changes, having the pair ve (‘he/she sees’) and veo (‘I see’). Therefore, we also need to pay attention to verbal morphology. If we do so, we can deduce the conjugation rules for the verb: -o = 1sg, = 3sg.\n\nThus, the final rules are:\n\nex:1:spanish\n\nWord order: S V O; Possessor – Possessed\n\nIf O = person, add a before it.\n\nVerb: 1sg = -o, 3sg =\n\nLet us now consider the following possible task:\n\n- There is an error in the following data. What is it?\n\nMi padre correo. | = | ‘My father runs.’\n\nSince we are told that there is an error in the sentence, we need to check the rules we have in order to see if any of them could justify this task. We remember that the -o ending of verbs is for 1sg subjects, but here the subject is 3sg, meaning there should be zero inflection on the verb, so the Spanish sentence should read Mi padre corre.\n\nIn the following chapters, each problem will be accompanied by a solution so that the reader can get accustomed to different ways of writing the solution."}
{"id":"book_method_c02_s01_p01","kind":"method","chapter":2,"chapter_title":"Writing systems","section":1,"section_title":"Introduction","topic":"writing systems and script decipherment","text":"Writing systems (or scripts) are collections of symbols and the rules for combining them in order to represent language. By some counts, there are now over 3,500 writing systems in the world. We will avoid the term letter to refer to these symbols since they might represent not only individual sounds (see chap-phonetics, Phonetics) but also syllables or even whole words. Therefore, the term character is preferred.","source_file":"chapters/02-WritingSystems.tex","source_line_start":5,"license":"CC-BY-4.0","retrieval_text":"Writing systems\nIntroduction\nwriting systems and script decipherment\nWriting systems (or scripts) are collections of symbols and the rules for combining them in order to represent language. By some counts, there are now over 3,500 writing systems in the world. We will avoid the term letter to refer to these symbols since they might represent not only individual sounds (see chap-phonetics, Phonetics) but also syllables or even whole words. Therefore, the term character is preferred."}
{"id":"book_method_c02_s02_p01","kind":"method","chapter":2,"chapter_title":"Writing systems","section":2,"section_title":"Pictographic and ideographic systems","topic":"writing systems and script decipherment","text":"Etymologically, the terms pictographic and ideographic are compounds formed from picto- (`picture'), ideo- (`idea'), and grafos (`writing'). In these systems, each character represents a word, an idea or a concept. Moreover, the characters are similar in appearance to the real-life representation of these concepts.\n\nAlthough there are notable differences between the picto- and ideographic systems, these are not very relevant when it comes to linguistics problems. For this reason, we will combine these two types of script in a single category.\n\nWe often encounter these kinds of systems in contexts where there is a need to convey some idea that is not specific to a particular language. Many such symbols are widely used across cultures, such as the P symbol used to represent a parking lot. A good example of a pictographic system is traffic and public signage (such as a crossed-out ice-cream cone to show that it is forbidden to eat food).\n\nIn linguistics problems these types of system are quite rare: usually, the relationship between the character and its meaning is so straightforward as to make the solution essentially effortless.","source_file":"chapters/02-WritingSystems.tex","source_line_start":9,"license":"CC-BY-4.0","retrieval_text":"Writing systems\nPictographic and ideographic systems\nwriting systems and script decipherment\nEtymologically, the terms pictographic and ideographic are compounds formed from picto- (`picture'), ideo- (`idea'), and grafos (`writing'). In these systems, each character represents a word, an idea or a concept. Moreover, the characters are similar in appearance to the real-life representation of these concepts.\n\nAlthough there are notable differences between the picto- and ideographic systems, these are not very relevant when it comes to linguistics problems. For this reason, we will combine these two types of script in a single category.\n\nWe often encounter these kinds of systems in contexts where there is a need to convey some idea that is not specific to a particular language. Many such symbols are widely used across cultures, such as the P symbol used to represent a parking lot. A good example of a pictographic system is traffic and public signage (such as a crossed-out ice-cream cone to show that it is forbidden to eat food).\n\nIn linguistics problems these types of system are quite rare: usually, the relationship between the character and its meaning is so straightforward as to make the solution essentially effortless."}
{"id":"book_method_c02_s03_p01","kind":"method","chapter":2,"chapter_title":"Writing systems","section":3,"section_title":"Logographic systems","topic":"writing systems and script decipherment","text":"In some ways, such scripts are highly similar to picto- and ideographic systems. The word logographic is made up of two etymons: logos- `word' and grafos- `writing': in these systems, at least in principle, each character represents a word. In most cases, the logographic systems have their origin in picto- or ideographic systems, but the characters have evolved over time, thus losing their similarity with the real-life representation of the words they designate. For instance, many Chinese characters have developed from pictographic representations to logographic ones. Thus, the character for ‘sun’, which used to be represented as something like ☉, is nowadays (in Modern Chinese) written as 日. Similarly, the character for ‘moon/month’, initially represented as something like ☽, has become 月 in Modern Chinese.","source_file":"chapters/02-WritingSystems.tex","source_line_start":19,"license":"CC-BY-4.0","retrieval_text":"Writing systems\nLogographic systems\nwriting systems and script decipherment\nIn some ways, such scripts are highly similar to picto- and ideographic systems. The word logographic is made up of two etymons: logos- `word' and grafos- `writing': in these systems, at least in principle, each character represents a word. In most cases, the logographic systems have their origin in picto- or ideographic systems, but the characters have evolved over time, thus losing their similarity with the real-life representation of the words they designate. For instance, many Chinese characters have developed from pictographic representations to logographic ones. Thus, the character for ‘sun’, which used to be represented as something like ☉, is nowadays (in Modern Chinese) written as 日. Similarly, the character for ‘moon/month’, initially represented as something like ☽, has become 月 in Modern Chinese."}
{"id":"book_method_c02_s04_p01","kind":"method","chapter":2,"chapter_title":"Writing systems","section":4,"section_title":"Syllabic systems (syllabaries)","topic":"writing systems and script decipherment","text":"Here, each character represents a syllable. Crucially, unlike some scripts that we will present below, each syllable is represented as a whole, without any clear relationship between characters denoting syllables that share some vowels or consonants.\n\nOne example of a syllabary is katakana, one of the writing systems in use in Japan.\n\nThe following table gives some examples of katakana characters. Each character represents a syllable, consisting of a consonant and a vowel. Conventionally, these are written in a table in which the vowel spans across columns and the consonant across rows:\n\n| a | i | u | e | o\n| ア | イ | ウ | エ | オ\nk | カ | キ | ク | ケ | コ\ns | サ | シ | ス | セ | ソ\nt | タ | チ | ツ | テ | ト\n\nWe can easily observe that the syllable is treated as a whole, and cannot be further divided into smaller components. For instance, the first line in the table corresponds to the absence of consonant (the character ウ represents the syllable u), but the other characters do not look anything like these symbols. Similarly, from knowing the characters for ka (カ), ki (キ), and si (シ) we cannot deduce the character for sa (サ).","source_file":"chapters/02-WritingSystems.tex","source_line_start":23,"license":"CC-BY-4.0","retrieval_text":"Writing systems\nSyllabic systems (syllabaries)\nwriting systems and script decipherment\nHere, each character represents a syllable. Crucially, unlike some scripts that we will present below, each syllable is represented as a whole, without any clear relationship between characters denoting syllables that share some vowels or consonants.\n\nOne example of a syllabary is katakana, one of the writing systems in use in Japan.\n\nThe following table gives some examples of katakana characters. Each character represents a syllable, consisting of a consonant and a vowel. Conventionally, these are written in a table in which the vowel spans across columns and the consonant across rows:\n\n| a | i | u | e | o\n| ア | イ | ウ | エ | オ\nk | カ | キ | ク | ケ | コ\ns | サ | シ | ス | セ | ソ\nt | タ | チ | ツ | テ | ト\n\nWe can easily observe that the syllable is treated as a whole, and cannot be further divided into smaller components. For instance, the first line in the table corresponds to the absence of consonant (the character ウ represents the syllable u), but the other characters do not look anything like these symbols. Similarly, from knowing the characters for ka (カ), ki (キ), and si (シ) we cannot deduce the character for sa (サ)."}
{"id":"book_method_c02_s05_p01","kind":"method","chapter":2,"chapter_title":"Writing systems","section":5,"section_title":"The Chinese writing system","topic":"writing systems and script decipherment","text":"There are a lot of misconceptions about the Chinese writing system. It is erroneous to consider it either pictographic or ideographic: Chinese characters are not “pictures” representing meanings directly. Neither is it entirely logographic. In fact, it is a logo-syllabic system (a logo-syllabary). Some characters are indeed purely logographic (as we mentioned before), but they make up a very small percentage of the total number of characters. Most of the characters are formed by what we call the rebus principle, where each character is formed by two components: a logographic part, approximately showing the meaning of that character (called the semantic component) and a syllabic part, giving clues about the pronunciation of that character (called the phonetic component).\n\nFor example, the character ma3 (马) is purely logographic and represents the word ‘horse’, while the character for ‘mother’ (妈, ma1) is logo-syllabic.To transcribe Chinese words, we use a modified version of the pinyin transcription system: the letters indicate the sounds and the digit indicates tone, a distinctive property of the entire syllable that can distinguish meaning (see Section sec:3.6). In Chinese, syllables with different tones differ in the level and the trajectory of the voice's pitch. It is formed from the semantic component (cf. the logographic 女 nü3 ‘woman’) and the phonetic component 马. Therefore, 妈 is a character whose meaning is related to ‘woman’ and whose pronunciation is similar to that of 马. Some more examples of logographic characters created by the rebus principle are:\n\n机 (ji1, ‘machine’) | = | 木 (mu4, ‘wood’) | + | 几 (ji3, ‘some’)\n唱 (chang4, ‘to sing’) | = | 口 (kou3, ‘mouth’) | + | 昌 (chang1, ‘prosperity’)","source_file":"chapters/02-WritingSystems.tex","source_line_start":44,"license":"CC-BY-4.0","retrieval_text":"Writing systems\nThe Chinese writing system\nwriting systems and script decipherment\nThere are a lot of misconceptions about the Chinese writing system. It is erroneous to consider it either pictographic or ideographic: Chinese characters are not “pictures” representing meanings directly. Neither is it entirely logographic. In fact, it is a logo-syllabic system (a logo-syllabary). Some characters are indeed purely logographic (as we mentioned before), but they make up a very small percentage of the total number of characters. Most of the characters are formed by what we call the rebus principle, where each character is formed by two components: a logographic part, approximately showing the meaning of that character (called the semantic component) and a syllabic part, giving clues about the pronunciation of that character (called the phonetic component).\n\nFor example, the character ma3 (马) is purely logographic and represents the word ‘horse’, while the character for ‘mother’ (妈, ma1) is logo-syllabic.To transcribe Chinese words, we use a modified version of the pinyin transcription system: the letters indicate the sounds and the digit indicates tone, a distinctive property of the entire syllable that can distinguish meaning (see Section sec:3.6). In Chinese, syllables with different tones differ in the level and the trajectory of the voice's pitch. It is formed from the semantic component (cf. the logographic 女 nü3 ‘woman’) and the phonetic component 马. Therefore, 妈 is a character whose meaning is related to ‘woman’ and whose pronunciation is similar to that of 马. Some more examples of logographic characters created by the rebus principle are:\n\n机 (ji1, ‘machine’) | = | 木 (mu4, ‘wood’) | + | 几 (ji3, ‘some’)\n唱 (chang4, ‘to sing’) | = | 口 (kou3, ‘mouth’) | + | 昌 (chang1, ‘prosperity’)"}
{"id":"book_method_c02_s06_p01","kind":"method","chapter":2,"chapter_title":"Writing systems","section":6,"section_title":"Alphabetic systems (alphabets)","topic":"writing systems and script decipherment","text":"These are the most common systems in Europe and each character represents either a consonant or a vowel. In other words, each character represents a sound. Examples of these scripts are the Latin alphabet (used to write languages such as English, Romanian, Spanish, Italian, German, Polish, Turkish), the Greek alphabet (used for writing Greek), the Cyrillic alphabet (used to write some Slavic languages such as Russian, Ukrainian, Belarusian, Bulgarian, as well as many languages spoken in Russia), the Georgian alphabet (used for the Georgian language), and the Armenian alphabet (used for the Armenian language).","source_file":"chapters/02-WritingSystems.tex","source_line_start":57,"license":"CC-BY-4.0","retrieval_text":"Writing systems\nAlphabetic systems (alphabets)\nwriting systems and script decipherment\nThese are the most common systems in Europe and each character represents either a consonant or a vowel. In other words, each character represents a sound. Examples of these scripts are the Latin alphabet (used to write languages such as English, Romanian, Spanish, Italian, German, Polish, Turkish), the Greek alphabet (used for writing Greek), the Cyrillic alphabet (used to write some Slavic languages such as Russian, Ukrainian, Belarusian, Bulgarian, as well as many languages spoken in Russia), the Georgian alphabet (used for the Georgian language), and the Armenian alphabet (used for the Armenian language)."}
{"id":"book_method_c02_s07_p01","kind":"method","chapter":2,"chapter_title":"Writing systems","section":7,"section_title":"Abjad systems","topic":"writing systems and script decipherment","text":"In these systems, each character represents a consonant, while vowels are not written at all or shown as diacritics (i.e., small marks attached to the character). In broad terms, this type of system can be compared to syllabic systems, in the sense that each “complete” character represents a combination of a consonant and a vowel; but, unlike syllabaries, each component is represented separately, so the character can be broken down into subparts. Some well-known abjads are used for Semitic languages, such as the Arabic and Hebrew scripts. Below we show some Arabic characters:\n| | u | ū | a | ā | i | ī\n1.2ema | 808080 | 1pt\"25CC \"064FOcFFL | \"06481pt\"25CC \"064FOcFFL | 1pt\"25CC \"064EOcFFL | \"06271pt\"25CC \"064EOcFFL | 1pt\"25CC -0.1em\"0650OcFFL | \"064A1pt\"25CC -0.1em\"0650OcFFL\nb1.2ema | ب | بُ | بُو | بَ | بَا | بِ | بِي\ns1.2ema | س | سُ | سُو | سَ | سَا | سِ | سِي\n\nThe dotted circle is used to show the placement of the diacritics for u, a, and i relative to the consonant character: in Arabic, the same diacritic is used to mark both a and i, but in the former case it is placed above the consonant, and in the latter case it goes below the character marking the consonant.","source_file":"chapters/02-WritingSystems.tex","source_line_start":60,"license":"CC-BY-4.0","retrieval_text":"Writing systems\nAbjad systems\nwriting systems and script decipherment\nIn these systems, each character represents a consonant, while vowels are not written at all or shown as diacritics (i.e., small marks attached to the character). In broad terms, this type of system can be compared to syllabic systems, in the sense that each “complete” character represents a combination of a consonant and a vowel; but, unlike syllabaries, each component is represented separately, so the character can be broken down into subparts. Some well-known abjads are used for Semitic languages, such as the Arabic and Hebrew scripts. Below we show some Arabic characters:\n| | u | ū | a | ā | i | ī\n1.2ema | 808080 | 1pt\"25CC \"064FOcFFL | \"06481pt\"25CC \"064FOcFFL | 1pt\"25CC \"064EOcFFL | \"06271pt\"25CC \"064EOcFFL | 1pt\"25CC -0.1em\"0650OcFFL | \"064A1pt\"25CC -0.1em\"0650OcFFL\nb1.2ema | ب | بُ | بُو | بَ | بَا | بِ | بِي\ns1.2ema | س | سُ | سُو | سَ | سَا | سِ | سِي\n\nThe dotted circle is used to show the placement of the diacritics for u, a, and i relative to the consonant character: in Arabic, the same diacritic is used to mark both a and i, but in the former case it is placed above the consonant, and in the latter case it goes below the character marking the consonant."}
{"id":"book_method_c02_s08_p01","kind":"method","chapter":2,"chapter_title":"Writing systems","section":8,"section_title":"Abugida systems","topic":"writing systems and script decipherment","text":"These systems are very similar to abjads, in the sense that each “complete” character represents a combination of a vowel and a consonant. However, in an abjad each character (without diacritics) represents a consonant and the diacritics, when used, append the corresponding vowels. In abugidas, a character without any diacritics or modifications represents a consonant followed by some vowel of the language (called the “default” or “inherent” vowel), and additional modifications are used to change that vowel to another one. Moreover, most abugidas will have some device to “delete” the vowel, so as to represent the consonant on its own.\n\nMany languages of South and South-East Asia are written using abugidas. Here are some Burmese characters:\n\n| a | aa | i | ii | u | uu | e | ai |\nA | \"25CC | ာ | ိ | ီ | ု | ူ | ေ | ဲ | ်\nkA | က | ကာ | ကိ | ကီ | ကု | ကူ | ကေ | ကဲ | က်\nsA | ဆ | ဆာ | ဆိ | ဆီ | ဆု | ဆူ | ဆေ | ဆဲ | ဆ်\n\nWe can observe that the default vowel is a (representing the basic form of each consonant). Adding a loop will change the vowel from a to aa (long a) while adding an L-like symbol below changes the vowel from a to u, etc.\n\nA very interesting abugida system is Cree Syllabics, used to write aboriginal languages of Canada. In this system, the vowel change is not shown by diacritics, but rather by rotating the character (by 90^, 180^ or 270^) or by mirroring it, depending on its shape. Below we show a table with some Cree characters.\n\nRotation | Indent | Mirroring\n1-5 7-11\n| a | e | i | o | | | a | e | i | o\n1-5 7-11\nA | ᐊ | ᐁ | ᐃ | ᐅ | | k | ᑲ | ᑫ | ᑭ | ᑯ\n1-5 7-11\npA | ᐸ | ᐯ | ᐱ | ᐳ | | m | ᒪ | ᒣ | ᒥ | ᒧ\n1-5 7-11\ntA | ᑕ | ᑌ | ᑎ | ᑐ | | n | ᓇ | ᓀ | ᓂ | ᓄ\n1-5 7-11\n\nWe can observe, conceptually, that the preference is for mirroring (right-hand table) since sometimes the same shape is used for two different consonants (k ᑲ, vertically, and n ᓇ, horizontally). The characters obtained by rotation are used only where mirroring is not possible due to symmetry reasons (a mirrored ᐁ is still ᐁ). Therefore, we can deduce that if a character cannot be mirrored (it has a symmetry axis), the vowel change will be represented by a clockwise 90^ rotation (vowels change in the sequence a i o e a). The change of vowel by mirroring is performed either vertically (a o and e i), or horizontally (a i and e o). The rotation and mirroring patterns of the Cree characters are shown below.\n\nRotation | Indent | Mirroring\n| ᐁ | | | ᑲ | ᑯ\n5-6\nᐅ | | ᐊ | | ᑭ | ᑫ\n| ᐃ |","source_file":"chapters/02-WritingSystems.tex","source_line_start":77,"license":"CC-BY-4.0","retrieval_text":"Writing systems\nAbugida systems\nwriting systems and script decipherment\nThese systems are very similar to abjads, in the sense that each “complete” character represents a combination of a vowel and a consonant. However, in an abjad each character (without diacritics) represents a consonant and the diacritics, when used, append the corresponding vowels. In abugidas, a character without any diacritics or modifications represents a consonant followed by some vowel of the language (called the “default” or “inherent” vowel), and additional modifications are used to change that vowel to another one. Moreover, most abugidas will have some device to “delete” the vowel, so as to represent the consonant on its own.\n\nMany languages of South and South-East Asia are written using abugidas. Here are some Burmese characters:\n\n| a | aa | i | ii | u | uu | e | ai |\nA | \"25CC | ာ | ိ | ီ | ု | ူ | ေ | ဲ | ်\nkA | က | ကာ | ကိ | ကီ | ကု | ကူ | ကေ | ကဲ | က်\nsA | ဆ | ဆာ | ဆိ | ဆီ | ဆု | ဆူ | ဆေ | ဆဲ | ဆ်\n\nWe can observe that the default vowel is a (representing the basic form of each consonant). Adding a loop will change the vowel from a to aa (long a) while adding an L-like symbol below changes the vowel from a to u, etc.\n\nA very interesting abugida system is Cree Syllabics, used to write aboriginal languages of Canada. In this system, the vowel change is not shown by diacritics, but rather by rotating the character (by 90^, 180^ or 270^) or by mirroring it, depending on its shape. Below we show a table with some Cree characters.\n\nRotation | Indent | Mirroring\n1-5 7-11\n| a | e | i | o | | | a | e | i | o\n1-5 7-11\nA | ᐊ | ᐁ | ᐃ | ᐅ | | k | ᑲ | ᑫ | ᑭ | ᑯ\n1-5 7-11\npA | ᐸ | ᐯ | ᐱ | ᐳ | | m | ᒪ | ᒣ | ᒥ | ᒧ\n1-5 7-11\ntA | ᑕ | ᑌ | ᑎ | ᑐ | | n | ᓇ | ᓀ | ᓂ | ᓄ\n1-5 7-11\n\nWe can observe, conceptually, that the preference is for mirroring (right-hand table) since sometimes the same shape is used for two different consonants (k ᑲ, vertically, and n ᓇ, horizontally). The characters obtained by rotation are used only where mirroring is not possible due to symmetry reasons (a mirrored ᐁ is still ᐁ). Therefore, we can deduce that if a character cannot be mirrored (it has a symmetry axis), the vowel change will be represented by a clockwise 90^ rotation (vowels change in the sequence a i o e a). The change of vowel by mirroring is performed either vertically (a o and e i), or horizontally (a i and e o). The rotation and mirroring patterns of the Cree characters are shown below.\n\nRotation | Indent | Mirroring\n| ᐁ | | | ᑲ | ᑯ\n5-6\nᐅ | | ᐊ | | ᑭ | ᑫ\n| ᐃ |"}
{"id":"book_method_c02_s09_p01","kind":"method","chapter":2,"chapter_title":"Writing systems","section":9,"section_title":"Featural systems","topic":"writing systems and script decipherment","text":"This is the last type of writing system and probably the least common. In this system, each character highlights the phonetic or phonological features of the sounds it designates. For example, in the Korean writing system (hangul), the characters corresponding to the sounds p, p*, ph, and m (ㅂ, ㅃ, ㅍ, ㅁ, respectively) are highly similar: all of them derive from a square. This is because all these sounds are bilabial (as we will see in the next chapter). When pronouncing them we use our lips (see the next chapter for more details on how to describe different sounds) and the square is meant to evoke the mouth seen from the front.\n\nThe property of being featural cuts across the other criteria we have used: the main characteristic of these systems is the fact that the properties of symbols are based on features of the sounds they represent, not the nature of the mapping between character and sound. For instance, the hangul writing system marks each character, whether vowel or consonant, individually, so it can also be considered an alphabet.\n\nThis is a syllabic type of system since each character represents a syllable. The only exception is the final character which is logographic (maybe even pictographic, considering that the symbol for `region' is composed of two symbols for `city', reflecting that a region is made up of more cities).\n\nA possible issue when solving the problem is the syllabification of the word Palaa. Normally, we would have been tempted to syllabify it as Pa-laa, assuming the aa represents a long vowel (similar to problem 2.1). Nevertheless, if we had done that, the region Palaa would have had only three characters (two characters for the syllables pa and laa and one logographic character for region). Since we do not have any three-character inscriptions, we chose to syllabify the word as Pa-la-a.","source_file":"chapters/02-WritingSystems.tex","source_line_start":118,"license":"CC-BY-4.0","retrieval_text":"Writing systems\nFeatural systems\nwriting systems and script decipherment\nThis is the last type of writing system and probably the least common. In this system, each character highlights the phonetic or phonological features of the sounds it designates. For example, in the Korean writing system (hangul), the characters corresponding to the sounds p, p*, ph, and m (ㅂ, ㅃ, ㅍ, ㅁ, respectively) are highly similar: all of them derive from a square. This is because all these sounds are bilabial (as we will see in the next chapter). When pronouncing them we use our lips (see the next chapter for more details on how to describe different sounds) and the square is meant to evoke the mouth seen from the front.\n\nThe property of being featural cuts across the other criteria we have used: the main characteristic of these systems is the fact that the properties of symbols are based on features of the sounds they represent, not the nature of the mapping between character and sound. For instance, the hangul writing system marks each character, whether vowel or consonant, individually, so it can also be considered an alphabet.\n\nThis is a syllabic type of system since each character represents a syllable. The only exception is the final character which is logographic (maybe even pictographic, considering that the symbol for `region' is composed of two symbols for `city', reflecting that a region is made up of more cities).\n\nA possible issue when solving the problem is the syllabification of the word Palaa. Normally, we would have been tempted to syllabify it as Pa-laa, assuming the aa represents a long vowel (similar to problem 2.1). Nevertheless, if we had done that, the region Palaa would have had only three characters (two characters for the syllables pa and laa and one logographic character for region). Since we do not have any three-character inscriptions, we chose to syllabify the word as Pa-la-a."}
{"id":"book_method_c02_s10_p01","kind":"method","chapter":2,"chapter_title":"Writing systems","section":10,"section_title":"Sign language","topic":"writing systems and script decipherment","text":"Sign languages are used by Deaf people and are based on movements and gestures rather than spoken sounds. Their grammar is entirely different from that of spoken languages, and they generally bear no relation to the languages of hearing communities: for instance, the United States and the United Kingdom share a majority spoken language (English), but their most widely used sign languages (American Sign Language/ASL, British Sign Language/BSL) are completely distinct and not mutually intelligible.\n\nIn addition to their own grammar, many sign languages have a system of finger spelling to represent the written forms of other languages. Thus, some gestures represent certain ideas or words (sometimes in an iconic way, thus having a semantic purpose), while others do not have any meaning and are representational of letters in written language, allowing the possibility to spell out items such as novel words or names.","source_file":"chapters/02-WritingSystems.tex","source_line_start":532,"license":"CC-BY-4.0","retrieval_text":"Writing systems\nSign language\nwriting systems and script decipherment\nSign languages are used by Deaf people and are based on movements and gestures rather than spoken sounds. Their grammar is entirely different from that of spoken languages, and they generally bear no relation to the languages of hearing communities: for instance, the United States and the United Kingdom share a majority spoken language (English), but their most widely used sign languages (American Sign Language/ASL, British Sign Language/BSL) are completely distinct and not mutually intelligible.\n\nIn addition to their own grammar, many sign languages have a system of finger spelling to represent the written forms of other languages. Thus, some gestures represent certain ideas or words (sometimes in an iconic way, thus having a semantic purpose), while others do not have any meaning and are representational of letters in written language, allowing the possibility to spell out items such as novel words or names."}
{"id":"book_method_c02_s11_p01","kind":"method","chapter":2,"chapter_title":"Writing systems","section":11,"section_title":"Braille alphabet","topic":"writing systems and script decipherment","text":"Braille is a writing system developed for blind people in which each character is represented by raised dots which the reader can feel with their finger tips. Each character is represented by a 32 grid\n|\n|\n|\n\nwhere each circle can be either empty (marking the lack of a raised dot) or full (marking a raised dot). This is just a way of representation in order to make it easier to identify. In reality, only the raised dots (full circles) are used. Just as in the case of sign languages, each country or language may develop their own Braille system, not necessarily mutually intelligible.","source_file":"chapters/02-WritingSystems.tex","source_line_start":538,"license":"CC-BY-4.0","retrieval_text":"Writing systems\nBraille alphabet\nwriting systems and script decipherment\nBraille is a writing system developed for blind people in which each character is represented by raised dots which the reader can feel with their finger tips. Each character is represented by a 32 grid\n|\n|\n|\n\nwhere each circle can be either empty (marking the lack of a raised dot) or full (marking a raised dot). This is just a way of representation in order to make it easier to identify. In reality, only the raised dots (full circles) are used. Just as in the case of sign languages, each country or language may develop their own Braille system, not necessarily mutually intelligible."}
{"id":"book_method_c03_s01_p01","kind":"method","chapter":3,"chapter_title":"Phonetics","section":1,"section_title":"Introduction","topic":"phonetics, stress, tone, and versification","text":"The field of phonetics is concerned with how speakers of different languages make the sounds of their speech (production, or articulation), and also with how they are heard by listeners (perception; this is sometimes known as auditory phonetics). Many linguistics problems will involve phenomena of phonology – for now, we can define these as changes of sounds that depend on the structure of words. We will see many examples of these in chap-phonology. Before we do so, it is useful to look at the different types of sound, and the ways in which they are classified, in order to understand these types of change.\n\nIn this chapter (and indeed throughout the book), we will use a notation system called the International Phonetic Alphabet (IPA). This is an alphabetic writing system, formed primarily of Latin characters and symbols derived therefrom and developed in order to make it possible to transcribe any word, from any spoken language, using this single set of conventions. Thus, each sound will have its own notation.\n\nWriting down the sounds of a language is referred to as (phonetic) transcription. To distinguish words that are transcribed from how they are represented in spelling, we often use square brackets (as in [tʰɹənsˈkɹɪp̚ʃn̩]) or slashes (as in /trænsˈkripʃən/), both corresponding to the spelling transcription. The precise difference between the two kinds of transcription shown here is not that important; very roughly, slashes are generally used for a kind of notation that only aims to differentiate the sounds found in that specific language (and thus omit some detail), while square brackets tend to contain quite narrow transcriptions, which contain a lot of specific information. For example, the sound usually written as /r/ can, in fact, be pronounced differently in different languages, and those differences will be reflected in narrower transcriptions: in English, it can be represented as [ɹ] (for instance, in many dialects of England), [ɻ] (in North America or Northern Ireland), or [ɾ] (in Scotland); in Romanian, it is generally [r], and in French or German it is usually [ʁ]. Narrow transcriptions aim to reflect this detail that depends on each language, but as often as not we only need to worry about the broad outlines of the system and will just use /r/ for all of these sounds, since most languages have only one type of r. This distinction rarely matters in practice for solving linguistics problems.","source_file":"chapters/03-phonetics.tex","source_line_start":5,"license":"CC-BY-4.0","retrieval_text":"Phonetics\nIntroduction\nphonetics, stress, tone, and versification\nThe field of phonetics is concerned with how speakers of different languages make the sounds of their speech (production, or articulation), and also with how they are heard by listeners (perception; this is sometimes known as auditory phonetics). Many linguistics problems will involve phenomena of phonology – for now, we can define these as changes of sounds that depend on the structure of words. We will see many examples of these in chap-phonology. Before we do so, it is useful to look at the different types of sound, and the ways in which they are classified, in order to understand these types of change.\n\nIn this chapter (and indeed throughout the book), we will use a notation system called the International Phonetic Alphabet (IPA). This is an alphabetic writing system, formed primarily of Latin characters and symbols derived therefrom and developed in order to make it possible to transcribe any word, from any spoken language, using this single set of conventions. Thus, each sound will have its own notation.\n\nWriting down the sounds of a language is referred to as (phonetic) transcription. To distinguish words that are transcribed from how they are represented in spelling, we often use square brackets (as in [tʰɹənsˈkɹɪp̚ʃn̩]) or slashes (as in /trænsˈkripʃən/), both corresponding to the spelling transcription. The precise difference between the two kinds of transcription shown here is not that important; very roughly, slashes are generally used for a kind of notation that only aims to differentiate the sounds found in that specific language (and thus omit some detail), while square brackets tend to contain quite narrow transcriptions, which contain a lot of specific information. For example, the sound usually written as /r/ can, in fact, be pronounced differently in different languages, and those differences will be reflected in narrower transcriptions: in English, it can be represented as [ɹ] (for instance, in many dialects of England), [ɻ] (in North America or Northern Ireland), or [ɾ] (in Scotland); in Romanian, it is generally [r], and in French or German it is usually [ʁ]. Narrow transcriptions aim to reflect this detail that depends on each language, but as often as not we only need to worry about the broad outlines of the system and will just use /r/ for all of these sounds, since most languages have only one type of r. This distinction rarely matters in practice for solving linguistics problems."}
{"id":"book_method_c03_s02_p01","kind":"method","chapter":3,"chapter_title":"Phonetics","section":2,"section_title":"Classification of sounds","topic":"phonetics, stress, tone, and versification","text":"Probably the most fundamental sound distinction that we need to keep in mind is the difference between vowels and consonants. The basic characteristic of vowels is that when we pronounce them, the air coming up from the lungs does not meet any obstacle (or, as we sometimes say, there is no constriction, either total or partial) in the vocal tract. Conversely, consonants are formed with some kind of obstacle, or stricture.\n\nConsonants\n\nWe can describe most consonants with reference to three main characteristics, which are referred to as place of articulation, manner of articulation, and voicing. (“Articulation” is just what we call the movements involved when we make the different sounds of speech.)\n\nPlace of articulation\n\nThis describes where in the vocal tract the main constriction is located, i.e., what organs are primarily involved in the articulation of each consonant. Going from front (the lips) to back (the back of the throat), we can identify the following places of articulation for consonants:A diagram of the vocal tract can be found on page 7 of the International Phonetic Association's Handbook of the International Phonetic Association, published by Cambridge University Press in 1999.\n\n- Labial, in which the lips are involved. In English, these are:\n\n- Bilabial consonants, produced using both lips (bi- = `two', labium = `lip' ‘two lips’): [b], [p], [m];\n\n- Labiodental consonants (labium, dental = `tooth' the lower lip and the upper teeth): [f], [v];\n\n- Coronal, pronounced by using the tip of the tongue:\n\n- (Inter)dental consonants (inter = `between', dental the tongue is placed between, or just on, the teeth) – [θ] (th in thin), [ð] (th in that);\n\n- Alveolar consonants (the tip of the tongue is placed on or near the alveolar ridge, the little protrusion just behind the upper teeth) – in English, [t], [d], [s], [z], [n], [l] are all usually alveolar, as is [r] in some languages other than English;In most linguistics problems, the sound [r] is treated as an alveolar sound, unless the footnotes state differently.\n\n- Post-alveolar consonants (post = ‘after’ the tongue is placed further behind than the alveolar ridge) – [ʃ] (sh in shop), [ʒ] (s in vision), [tʃ] (ch in church), [dʒ] (j in jam); in many varieties of English, [ɹ] is also post-alveolar;\n\n- Dorsal consonants, articulated primarily using the blade of the tongue:\n\n- Palatal consonants (the tongue is placed against the hard palate of the oral cavity): one example from English is [j] (note that in the IPA this symbol refers to the y of yellow, not the j of jam!). Other palatal consonants are [c] (similar to cc in accute, but not as the c in cat), [ɟ] (similar to g in geese), [ɲ] (ni in onion), [ʎ] (as in Italian figlio, or in English million);\n\n- Velar consonants (the back of the tongue is placed against the soft palate, also known as the velum) – [k] (c in cat), [ɡ], [ŋ] (ng in king), [x] (as in Scottish loch or German Bach);\n\n- Glottal consonants, which generally lack any constriction in the mouth, but some noise is created as air passes through the vocal folds (the glottis is the gap between the two vocal folds) – [h] and the glottal stop, written [ʔ] and heard in uh-oh, or, for some speakers, in the middle of words like butter. (Pronouncing butter with a glottal stop is sometimes called “dropping one's t's”, but that is clearly wrong: the t isn't dropped, it's just pronounced differently!)","source_file":"chapters/03-phonetics.tex","source_line_start":13,"license":"CC-BY-4.0","retrieval_text":"Phonetics\nClassification of sounds\nphonetics, stress, tone, and versification\nProbably the most fundamental sound distinction that we need to keep in mind is the difference between vowels and consonants. The basic characteristic of vowels is that when we pronounce them, the air coming up from the lungs does not meet any obstacle (or, as we sometimes say, there is no constriction, either total or partial) in the vocal tract. Conversely, consonants are formed with some kind of obstacle, or stricture.\n\nConsonants\n\nWe can describe most consonants with reference to three main characteristics, which are referred to as place of articulation, manner of articulation, and voicing. (“Articulation” is just what we call the movements involved when we make the different sounds of speech.)\n\nPlace of articulation\n\nThis describes where in the vocal tract the main constriction is located, i.e., what organs are primarily involved in the articulation of each consonant. Going from front (the lips) to back (the back of the throat), we can identify the following places of articulation for consonants:A diagram of the vocal tract can be found on page 7 of the International Phonetic Association's Handbook of the International Phonetic Association, published by Cambridge University Press in 1999.\n\n- Labial, in which the lips are involved. In English, these are:\n\n- Bilabial consonants, produced using both lips (bi- = `two', labium = `lip' ‘two lips’): [b], [p], [m];\n\n- Labiodental consonants (labium, dental = `tooth' the lower lip and the upper teeth): [f], [v];\n\n- Coronal, pronounced by using the tip of the tongue:\n\n- (Inter)dental consonants (inter = `between', dental the tongue is placed between, or just on, the teeth) – [θ] (th in thin), [ð] (th in that);\n\n- Alveolar consonants (the tip of the tongue is placed on or near the alveolar ridge, the little protrusion just behind the upper teeth) – in English, [t], [d], [s], [z], [n], [l] are all usually alveolar, as is [r] in some languages other than English;In most linguistics problems, the sound [r] is treated as an alveolar sound, unless the footnotes state differently.\n\n- Post-alveolar consonants (post = ‘after’ the tongue is placed further behind than the alveolar ridge) – [ʃ] (sh in shop), [ʒ] (s in vision), [tʃ] (ch in church), [dʒ] (j in jam); in many varieties of English, [ɹ] is also post-alveolar;\n\n- Dorsal consonants, articulated primarily using the blade of the tongue:\n\n- Palatal consonants (the tongue is placed against the hard palate of the oral cavity): one example from English is [j] (note that in the IPA this symbol refers to the y of yellow, not the j of jam!). Other palatal consonants are [c] (similar to cc in accute, but not as the c in cat), [ɟ] (similar to g in geese), [ɲ] (ni in onion), [ʎ] (as in Italian figlio, or in English million);\n\n- Velar consonants (the back of the tongue is placed against the soft palate, also known as the velum) – [k] (c in cat), [ɡ], [ŋ] (ng in king), [x] (as in Scottish loch or German Bach);\n\n- Glottal consonants, which generally lack any constriction in the mouth, but some noise is created as air passes through the vocal folds (the glottis is the gap between the two vocal folds) – [h] and the glottal stop, written [ʔ] and heard in uh-oh, or, for some speakers, in the middle of words like butter. (Pronouncing butter with a glottal stop is sometimes called “dropping one's t's”, but that is clearly wrong: the t isn't dropped, it's just pronounced differently!)"}
{"id":"book_method_c03_s02_p02","kind":"method","chapter":3,"chapter_title":"Phonetics","section":2,"section_title":"Classification of sounds","topic":"phonetics, stress, tone, and versification","text":"- Glottal consonants, which generally lack any constriction in the mouth, but some noise is created as air passes through the vocal folds (the glottis is the gap between the two vocal folds) – [h] and the glottal stop, written [ʔ] and heard in uh-oh, or, for some speakers, in the middle of words like butter. (Pronouncing butter with a glottal stop is sometimes called “dropping one's t's”, but that is clearly wrong: the t isn't dropped, it's just pronounced differently!)\n\nThese are the main places of articulation we encounter, but in the world's languages, there are quite a few more (alveolo-palatal, retroflex, uvular, pharyngeal, epiglottal, and so on). In practice, though, if these kinds of sounds are featured in a linguistics problem, they will likely be described in the note at the end: yet another reason to read these notes carefully. If the consonants are described in apparently unnecessary detail, this might be a clue that the information is important!\n\nManner of articulation\n\nManner refers to the precise way in which the articulators move and the stricture is formed. Here is how the consonants can be classified in terms of manner:\n\n- Plosive consonants, also known as occlusives or stops, formed with a total closure of the vocal tract, followed by a sudden release, similar to an explosion. Examples of these consonants are: [p], [b], [t], [d], [k], [ɡ];\n\n- Fricative consonants: there is a partial stricture of the vocal tract that leaves a very narrow opening. As a result, the constant airflow through the opening produces turbulent noise. Examples include [f], [v], [s], [z], [θ], [ð] [x], [ɣ];\n\n- Affricate consonants are a combination of a plosive and a fricative: at the beginning of their articulation, the closure of the vocal tract is total, but the release is gradual, causing a fricative-like flow of air. In the IPA, they are notated by the symbol for the stop followed by the corresponding fricative, sometimes joined by an arc: [t͡s], [d͡z], [t͡ʃ], [d͡ʒ].The addition of the arc is necessary because some languages distinguish between the affricate and the corresponding stop-fricative sequence (as in Polish trzy [tʃɨ] ‘three’ but czy [t͡ʃɨ] ‘whether’). Nevertheless, this rarely happens in linguistics problems, and you should not worry about the presence or absence of the arc.\nNote also the letter c, which in many languages' orthographies is used for the affricate [t͡s], but has a different meaning in the IPA!\n\n- Nasal consonants: in this case, the oral tract is closed (as it is for stops), but the airflow is released through the nose. Most IPA symbols for nasals resemble the letter n, e.g.: [n], [m], [ŋ], [ɳ];\n\n- Liquid consonants. This is an umbrella term for many different manners of articulation, but as far as the linguistics problems are concerned, we need not go into any more details. This category includes the consonants [l] and [r], as well as, similar to the nasals, most consonants whose IPA symbols resemble l and r (e.g., [ɫ], [ʟ], [ʎ], [ɺ], [ɽ], [ʀ]).\n\n- Glides (also known as semivowels). There is still partial closure of the vocal tract, but it is not as narrow as for fricatives: these consonants are in between fricative consonants and vowels. Commonly encountered glides are [w] (w in week) and [j] (y in year).\n\n- Technically, [w] is what is known as a “labial-velar” glide because it is articulated using both the back of the tongue and the lips. In real problems that you might encounter, it might behave both as a velar and as a labial, so we have put it in both columns in Table tab:conschart.\n\nThere are also umbrella terms that can be useful with respect to the manner of articulation. The cover term for plosives, fricatives and affricates is obstruent, while all the other consonants (nasals, liquids, and glides) can be together referred to as sonorants.","source_file":"chapters/03-phonetics.tex","source_line_start":13,"license":"CC-BY-4.0","retrieval_text":"Phonetics\nClassification of sounds\nphonetics, stress, tone, and versification\n- Glottal consonants, which generally lack any constriction in the mouth, but some noise is created as air passes through the vocal folds (the glottis is the gap between the two vocal folds) – [h] and the glottal stop, written [ʔ] and heard in uh-oh, or, for some speakers, in the middle of words like butter. (Pronouncing butter with a glottal stop is sometimes called “dropping one's t's”, but that is clearly wrong: the t isn't dropped, it's just pronounced differently!)\n\nThese are the main places of articulation we encounter, but in the world's languages, there are quite a few more (alveolo-palatal, retroflex, uvular, pharyngeal, epiglottal, and so on). In practice, though, if these kinds of sounds are featured in a linguistics problem, they will likely be described in the note at the end: yet another reason to read these notes carefully. If the consonants are described in apparently unnecessary detail, this might be a clue that the information is important!\n\nManner of articulation\n\nManner refers to the precise way in which the articulators move and the stricture is formed. Here is how the consonants can be classified in terms of manner:\n\n- Plosive consonants, also known as occlusives or stops, formed with a total closure of the vocal tract, followed by a sudden release, similar to an explosion. Examples of these consonants are: [p], [b], [t], [d], [k], [ɡ];\n\n- Fricative consonants: there is a partial stricture of the vocal tract that leaves a very narrow opening. As a result, the constant airflow through the opening produces turbulent noise. Examples include [f], [v], [s], [z], [θ], [ð] [x], [ɣ];\n\n- Affricate consonants are a combination of a plosive and a fricative: at the beginning of their articulation, the closure of the vocal tract is total, but the release is gradual, causing a fricative-like flow of air. In the IPA, they are notated by the symbol for the stop followed by the corresponding fricative, sometimes joined by an arc: [t͡s], [d͡z], [t͡ʃ], [d͡ʒ].The addition of the arc is necessary because some languages distinguish between the affricate and the corresponding stop-fricative sequence (as in Polish trzy [tʃɨ] ‘three’ but czy [t͡ʃɨ] ‘whether’). Nevertheless, this rarely happens in linguistics problems, and you should not worry about the presence or absence of the arc.\nNote also the letter c, which in many languages' orthographies is used for the affricate [t͡s], but has a different meaning in the IPA!\n\n- Nasal consonants: in this case, the oral tract is closed (as it is for stops), but the airflow is released through the nose. Most IPA symbols for nasals resemble the letter n, e.g.: [n], [m], [ŋ], [ɳ];\n\n- Liquid consonants. This is an umbrella term for many different manners of articulation, but as far as the linguistics problems are concerned, we need not go into any more details. This category includes the consonants [l] and [r], as well as, similar to the nasals, most consonants whose IPA symbols resemble l and r (e.g., [ɫ], [ʟ], [ʎ], [ɺ], [ɽ], [ʀ]).\n\n- Glides (also known as semivowels). There is still partial closure of the vocal tract, but it is not as narrow as for fricatives: these consonants are in between fricative consonants and vowels. Commonly encountered glides are [w] (w in week) and [j] (y in year).\n\n- Technically, [w] is what is known as a “labial-velar” glide because it is articulated using both the back of the tongue and the lips. In real problems that you might encounter, it might behave both as a velar and as a labial, so we have put it in both columns in Table tab:conschart.\n\nThere are also umbrella terms that can be useful with respect to the manner of articulation. The cover term for plosives, fricatives and affricates is obstruent, while all the other consonants (nasals, liquids, and glides) can be together referred to as sonorants."}
{"id":"book_method_c03_s02_p03","kind":"method","chapter":3,"chapter_title":"Phonetics","section":2,"section_title":"Classification of sounds","topic":"phonetics, stress, tone, and versification","text":"There are also umbrella terms that can be useful with respect to the manner of articulation. The cover term for plosives, fricatives and affricates is obstruent, while all the other consonants (nasals, liquids, and glides) can be together referred to as sonorants.\n\nMoreover, there is a useful term that combines a manner of articulation with a place of articulation: alveolar and post-alveolar fricatives and affricates can be called sibilants.\n\nVoicing\n\nVoicing refers to the involvement of the vocal folds in the articulation of the consonants. A common distinction is between voiceless (vocal folds are not involved) and voiced (vocal folds are vibrating) sounds. Vowels are almost always voiced, but this parameter can make a difference for consonants. Specifically:\n\n- sonorants (i.e., nasals, liquids, and glides) are almost always voiced;\n\n- stops, fricatives, and affricates can be either voiceless ([p], [s], [ts], [k]) or voiced ([b], [z], [dz], [ɡ]).\n\nConsidering all of the above, we can lay out the consonants in a table summarising all these properties (Table tab:conschart). Across columns, we will write the places of articulation (left to right from anterior to posterior), while rows show the different manners of articulation. To show voicing, we can use alignment within the table cell: the voiceless consonant is on the left and the voiced one is on the right. For nasals, liquids and glides (for which there is usually no voiceless correspondent), the symbol will be placed in the centre.\n\n=.9\n| | labial | coronal | dorsal\n(lr)3-6(lr)7-12(lr)13-18\n| | bilabial | labio- | (inter) | alveolar | post- | palatal | velar | glottal\n| | | | dental | dental | | | alveolar | | | | | |\nstops | p | b | | | | | t | d | | | c | ɟ | k | ɡ | ʔ |\nfricatives | | | f | v | θ | ð | s | z | ʃ | ʒ | | | x | ɣ | h |\naffricates | | | | | | | ts | dz | tʃ | dʒ | | | | | |\nnasals | m | | | | | n | | | ɲ | ŋ | |\nliquids |\n| lateral | | | | | | | l | | | | | | | |\n| rhotic | | | | | | | r | | | | | | | |\nglides | (w) | | | | | | | | | j | (w) | |\n\nCaption: Consonants\n\nWe have also included in this table some other consonants that we have not discussed, but are commonly featured in linguistics problems: the glottal stop ([ʔ]), as well as the velar fricatives ([x] and [ɣ]). We can describe every consonant with reference to the properties we've just outlined, for example:\n\n- [f] = voiceless labiodental fricative\n\n- [m] = bilabial nasal\n\n- [ɡ] = voiced velar stop\n\nIn doing this, the usual order is voicing – place – manner.\n\nOther characteristics of consonants\n\nBesides the three main characteristics mentioned above, consonants can also have other features, such as:","source_file":"chapters/03-phonetics.tex","source_line_start":13,"license":"CC-BY-4.0","retrieval_text":"Phonetics\nClassification of sounds\nphonetics, stress, tone, and versification\nThere are also umbrella terms that can be useful with respect to the manner of articulation. The cover term for plosives, fricatives and affricates is obstruent, while all the other consonants (nasals, liquids, and glides) can be together referred to as sonorants.\n\nMoreover, there is a useful term that combines a manner of articulation with a place of articulation: alveolar and post-alveolar fricatives and affricates can be called sibilants.\n\nVoicing\n\nVoicing refers to the involvement of the vocal folds in the articulation of the consonants. A common distinction is between voiceless (vocal folds are not involved) and voiced (vocal folds are vibrating) sounds. Vowels are almost always voiced, but this parameter can make a difference for consonants. Specifically:\n\n- sonorants (i.e., nasals, liquids, and glides) are almost always voiced;\n\n- stops, fricatives, and affricates can be either voiceless ([p], [s], [ts], [k]) or voiced ([b], [z], [dz], [ɡ]).\n\nConsidering all of the above, we can lay out the consonants in a table summarising all these properties (Table tab:conschart). Across columns, we will write the places of articulation (left to right from anterior to posterior), while rows show the different manners of articulation. To show voicing, we can use alignment within the table cell: the voiceless consonant is on the left and the voiced one is on the right. For nasals, liquids and glides (for which there is usually no voiceless correspondent), the symbol will be placed in the centre.\n\n=.9\n| | labial | coronal | dorsal\n(lr)3-6(lr)7-12(lr)13-18\n| | bilabial | labio- | (inter) | alveolar | post- | palatal | velar | glottal\n| | | | dental | dental | | | alveolar | | | | | |\nstops | p | b | | | | | t | d | | | c | ɟ | k | ɡ | ʔ |\nfricatives | | | f | v | θ | ð | s | z | ʃ | ʒ | | | x | ɣ | h |\naffricates | | | | | | | ts | dz | tʃ | dʒ | | | | | |\nnasals | m | | | | | n | | | ɲ | ŋ | |\nliquids |\n| lateral | | | | | | | l | | | | | | | |\n| rhotic | | | | | | | r | | | | | | | |\nglides | (w) | | | | | | | | | j | (w) | |\n\nCaption: Consonants\n\nWe have also included in this table some other consonants that we have not discussed, but are commonly featured in linguistics problems: the glottal stop ([ʔ]), as well as the velar fricatives ([x] and [ɣ]). We can describe every consonant with reference to the properties we've just outlined, for example:\n\n- [f] = voiceless labiodental fricative\n\n- [m] = bilabial nasal\n\n- [ɡ] = voiced velar stop\n\nIn doing this, the usual order is voicing – place – manner.\n\nOther characteristics of consonants\n\nBesides the three main characteristics mentioned above, consonants can also have other features, such as:"}
{"id":"book_method_c03_s02_p04","kind":"method","chapter":3,"chapter_title":"Phonetics","section":2,"section_title":"Classification of sounds","topic":"phonetics, stress, tone, and versification","text":"Besides the three main characteristics mentioned above, consonants can also have other features, such as:\n\n- Aspiration – some consonants, especially stops, can be aspirated, i.e., pronounced with a little puff of air. This feature is marked by a superscript letter h after the consonant symbol. Thus, an aspirated, voiceless, bilabial stop can be written as [pʰ], since the sound is similar to the pronunciation of the consonant, followed by an [h]. Although in most languages of Europe consonants with the same place and manner of articulation are usually differentiated by voicing ([p] vs. [b], [f] vs. [v], etc.), many others differentiate these sounds based on aspiration; there are languages that have a three-way distinction between aspirated voiceless, unaspirated voiceless and unaspirated voiced versions of the same stop. For example, Mandarin Chinese does not have any voiced stops, but it has aspirated and unaspirated stops. Although the standard Chinese transliteration system (pinyin) uses the letters p and b, in reality, they correspond to the sounds [pʰ] and [p], respectively.\n\n- Labialisation is similar to aspiration in the sense that consonants generally are non-labialised, but can exist in a labialised variant (called labialised consonants). An alternative term is rounding, which is also used for vowels (see below). Labialised consonants are pronounced with rounded lips, and they are marked in the IPA by a superscript [w] symbol. Thus, a labialised voiced velar stop is written as [ɡw].\n\n- Palatalisation is another optional feature, similar to labialisation: in palatalised consonants, the back of the tongue is raised upwards, towards the hard palate. It is marked by a superscript [j], though palatalisation can sometimes be indicated by an added apostrophe, e.g. [t'].\n\nThere are many other signs and symbols that can be added to a consonant to mark different alterations, but, if they are featured in a linguistics problem and are relevant to solving that problem, they will be described in the footnotes.\n\nBesides palatalisation, there are other symbols that show the shifting of the tongue, e.g., velarisation (superscript [ɣ]) and pharyngealisation (superscript [ʕ]).\n\nFor example, while nasals and liquids are usually voiced, some languages also possess voiceless sonorants, for which there are no dedicated IPA symbols. To mark the voicelessness of a sonorant consonant or a vowel, a small ring can be placed under (in some cases, above) the corresponding symbol. For example, a voiceless alveolar nasal is written as [n̥].\n\nVowels\n\nVowels are produced without any constriction of the vocal tract, but with a subtle narrowing of the oral cavity by the tongue. The main features relevant to them are backness, height (aperture) and roundness.\n\nBackness\n\nBackness refers to the position of the tongue during articulation relative to the back of the mouth: to some extent, it is comparable with the place of articulation of consonants (and indeed in some languages the two might interact). Nevertheless, unlike consonants where the place of articulation is discrete (it has a small number of fixed values), backness is a more continuous parameter. On a very basic level, vowels are classified into front, central, and back, but they can also be subdivided further: between the front and central vowels there are near-front vowels and between central and back there are near-back vowels. For simplicity, we will only consider the three basic values of backness, as follows:\n\n- front vowels: [i] (ee in free), [e] (e in Spanish), [ø] (ö in German or Turkish or eu in the French word peu), [y] (ü in German or Turkish, or u in the French pu), [ɛ] (e in hen), [a] (a front a is traditional in French words like patte);","source_file":"chapters/03-phonetics.tex","source_line_start":13,"license":"CC-BY-4.0","retrieval_text":"Phonetics\nClassification of sounds\nphonetics, stress, tone, and versification\nBesides the three main characteristics mentioned above, consonants can also have other features, such as:\n\n- Aspiration – some consonants, especially stops, can be aspirated, i.e., pronounced with a little puff of air. This feature is marked by a superscript letter h after the consonant symbol. Thus, an aspirated, voiceless, bilabial stop can be written as [pʰ], since the sound is similar to the pronunciation of the consonant, followed by an [h]. Although in most languages of Europe consonants with the same place and manner of articulation are usually differentiated by voicing ([p] vs. [b], [f] vs. [v], etc.), many others differentiate these sounds based on aspiration; there are languages that have a three-way distinction between aspirated voiceless, unaspirated voiceless and unaspirated voiced versions of the same stop. For example, Mandarin Chinese does not have any voiced stops, but it has aspirated and unaspirated stops. Although the standard Chinese transliteration system (pinyin) uses the letters p and b, in reality, they correspond to the sounds [pʰ] and [p], respectively.\n\n- Labialisation is similar to aspiration in the sense that consonants generally are non-labialised, but can exist in a labialised variant (called labialised consonants). An alternative term is rounding, which is also used for vowels (see below). Labialised consonants are pronounced with rounded lips, and they are marked in the IPA by a superscript [w] symbol. Thus, a labialised voiced velar stop is written as [ɡw].\n\n- Palatalisation is another optional feature, similar to labialisation: in palatalised consonants, the back of the tongue is raised upwards, towards the hard palate. It is marked by a superscript [j], though palatalisation can sometimes be indicated by an added apostrophe, e.g. [t'].\n\nThere are many other signs and symbols that can be added to a consonant to mark different alterations, but, if they are featured in a linguistics problem and are relevant to solving that problem, they will be described in the footnotes.\n\nBesides palatalisation, there are other symbols that show the shifting of the tongue, e.g., velarisation (superscript [ɣ]) and pharyngealisation (superscript [ʕ]).\n\nFor example, while nasals and liquids are usually voiced, some languages also possess voiceless sonorants, for which there are no dedicated IPA symbols. To mark the voicelessness of a sonorant consonant or a vowel, a small ring can be placed under (in some cases, above) the corresponding symbol. For example, a voiceless alveolar nasal is written as [n̥].\n\nVowels\n\nVowels are produced without any constriction of the vocal tract, but with a subtle narrowing of the oral cavity by the tongue. The main features relevant to them are backness, height (aperture) and roundness.\n\nBackness\n\nBackness refers to the position of the tongue during articulation relative to the back of the mouth: to some extent, it is comparable with the place of articulation of consonants (and indeed in some languages the two might interact). Nevertheless, unlike consonants where the place of articulation is discrete (it has a small number of fixed values), backness is a more continuous parameter. On a very basic level, vowels are classified into front, central, and back, but they can also be subdivided further: between the front and central vowels there are near-front vowels and between central and back there are near-back vowels. For simplicity, we will only consider the three basic values of backness, as follows:\n\n- front vowels: [i] (ee in free), [e] (e in Spanish), [ø] (ö in German or Turkish or eu in the French word peu), [y] (ü in German or Turkish, or u in the French pu), [ɛ] (e in hen), [a] (a front a is traditional in French words like patte);"}
{"id":"book_method_c03_s02_p05","kind":"method","chapter":3,"chapter_title":"Phonetics","section":2,"section_title":"Classification of sounds","topic":"phonetics, stress, tone, and versification","text":"- front vowels: [i] (ee in free), [e] (e in Spanish), [ø] (ö in German or Turkish or eu in the French word peu), [y] (ü in German or Turkish, or u in the French pu), [ɛ] (e in hen), [a] (a front a is traditional in French words like patte);\n\n- central vowels: [ɐ] (u in nut), [ə] (a in above), [ɨ] (ы in Russian). There is no dedicated symbol in the IPA for a central low vowel (as in Spanish a): technically, IPA [a] is front, but in a linguistics problem <a> will often have the “European” value of the central low vowel;\n\n- back vowels: [ɔ] (a in call),[o] (oa in boat, in Scottish English), [u] (oo in boot);\n\nHeight\n\nHeight (or aperture) refers to the vertical position of the tongue and the jaw when articulating the sound. Similar to backness, it is a continuous parameter. Generally, we talk about high (close), mid, and low (open) vowels. These can be further subdivided into near-close (near-high), close-mid (high-mid), open-mid (low-mid), near-open (near-low). Again, for simplicity, we will only consider the three main values, as follows:\n\n- low vowels: [a], [ɐ], [æ];\n\n- mid vowels: [e], [ɛ], [ɔ], [ø], [ə], [o];\n\n- high-mid vowels: [e], [o], [ø];\n\n- low-mid vowels: [ɛ], [ɔ];\n\n- [ə] (called schwa) is somewhat special: it is central in backness and mid in height (it represents the resting state of the vocal tract). It can pattern in different ways in different languages and is often (but not always) found only in unstressed syllables.\n\n- high vowels: [i], [y], [ɨ], [u].\n\nRoundness\n\nThe last main characteristic of vowels is roundness. This, similar to the voicing of consonants, is a binary parameter. We can differentiate:\n\n- rounded vowels – for which the lips are rounded during articulation: [o], [u], [y], [ø], [ɔ];\n\n- unrounded vowels – for which the lips are not rounded during articulation: [ɐ], [æ], [ə], [ɨ], [e], [ɛ], [i].\n\nWe can present all these characteristics as in Figure fig:vowel-space. This time, since both the backness and height are continuous parameters, it is not a table, but a diagram showing the vowel space.\n\nvowel/.style=circle,fill=black,minimum size=0.75ex,inner sep=0pt, outer sep=3pt,\nevery label/.style=font=,fill=white,outer sep=0pt, inner sep=0pt,\ntitlenode/.style=font=,anchor=west\n\nat (2,0.75) Front;\nat (4,0.75) Central;\nat (6,0.75) Back;\n\nat (-1,0) Close;\nat (2,0) [vowel, label=right:y, label=left:i] (close-front) ;\nat (4,0) [vowel, label=right:ʉ, label=left:ɨ] (close-central) ;\nat (6,0) [vowel, label=right:u, label=left:ɯ] (close-back) ;\n\nat (-1,-1) Close-mid;\nat (2.5,-1) [vowel, label=right:ø, label=left:e] (cmid-front);\nat (6,-1) [vowel, label=right:o, label=left:ɤ] (cmid-back);\n\nat (-1,-2) Open-mid;\nat (3,-2) [vowel, label=right:œ, label=left:ɛ] (omid-front) ;\nat (6,-2) [vowel, label=right:ɔ, label=left:ʌ] (omid-back) ;\n\nat (-1,-3) Open;\nat (3.5,-3) [vowel, label=left:a] (open-front) ;\nat (6,-3) [vowel, label=right:ɒ, label=left:ɑ] (open-back) ;\n\nat (4.75,-3) (lower-border) ;\nat (4.6,-2.5) ɐ;\nat (4.4,-1.5) ə;\n\n(close-front) – (close-central);\n(close-central) – (close-back);\n(close-back) – (open-back);\n(open-back) – (open-front);\n(open-front) – (close-front) node [very near start, left] æ;\n\n(cmid-front) – (cmid-back);\n(omid-front) – (omid-back);\n\n(close-central) – (lower-border);\n\nCaption: Vowel space\n\nIn this diagram, the two dimensions are height (y-axis) and backness (x-axis), while roundness is determined by the vowel position relative to the reference point: symbols to the left of the dot refer to unrounded vowels and those on the right are rounded.\n\nOther characteristics of vowels\n\nThere are some other features of vowels, usually marked by diacritics or superscript symbols. For example:\n\n- Tongue root position – can have three values:\n\n- neutral = the default tongue root position;\n\n- advanced = when pronouncing the vowel, the tongue root is slightly shifted towards the front. These vowels are called ATR (advanced tongue root) and are marked in the IPA by the diacritic ◌̘;\n\n- retracted = the tongue root is shifted towards the back (retracted). These vowels are called RTR (retracted tongue root) and are marked by ◌̙;","source_file":"chapters/03-phonetics.tex","source_line_start":13,"license":"CC-BY-4.0","retrieval_text":"Phonetics\nClassification of sounds\nphonetics, stress, tone, and versification\n- front vowels: [i] (ee in free), [e] (e in Spanish), [ø] (ö in German or Turkish or eu in the French word peu), [y] (ü in German or Turkish, or u in the French pu), [ɛ] (e in hen), [a] (a front a is traditional in French words like patte);\n\n- central vowels: [ɐ] (u in nut), [ə] (a in above), [ɨ] (ы in Russian). There is no dedicated symbol in the IPA for a central low vowel (as in Spanish a): technically, IPA [a] is front, but in a linguistics problem <a> will often have the “European” value of the central low vowel;\n\n- back vowels: [ɔ] (a in call),[o] (oa in boat, in Scottish English), [u] (oo in boot);\n\nHeight\n\nHeight (or aperture) refers to the vertical position of the tongue and the jaw when articulating the sound. Similar to backness, it is a continuous parameter. Generally, we talk about high (close), mid, and low (open) vowels. These can be further subdivided into near-close (near-high), close-mid (high-mid), open-mid (low-mid), near-open (near-low). Again, for simplicity, we will only consider the three main values, as follows:\n\n- low vowels: [a], [ɐ], [æ];\n\n- mid vowels: [e], [ɛ], [ɔ], [ø], [ə], [o];\n\n- high-mid vowels: [e], [o], [ø];\n\n- low-mid vowels: [ɛ], [ɔ];\n\n- [ə] (called schwa) is somewhat special: it is central in backness and mid in height (it represents the resting state of the vocal tract). It can pattern in different ways in different languages and is often (but not always) found only in unstressed syllables.\n\n- high vowels: [i], [y], [ɨ], [u].\n\nRoundness\n\nThe last main characteristic of vowels is roundness. This, similar to the voicing of consonants, is a binary parameter. We can differentiate:\n\n- rounded vowels – for which the lips are rounded during articulation: [o], [u], [y], [ø], [ɔ];\n\n- unrounded vowels – for which the lips are not rounded during articulation: [ɐ], [æ], [ə], [ɨ], [e], [ɛ], [i].\n\nWe can present all these characteristics as in Figure fig:vowel-space. This time, since both the backness and height are continuous parameters, it is not a table, but a diagram showing the vowel space.\n\nvowel/.style=circle,fill=black,minimum size=0.75ex,inner sep=0pt, outer sep=3pt,\nevery label/.style=font=,fill=white,outer sep=0pt, inner sep=0pt,\ntitlenode/.style=font=,anchor=west\n\nat (2,0.75) Front;\nat (4,0.75) Central;\nat (6,0.75) Back;\n\nat (-1,0) Close;\nat (2,0) [vowel, label=right:y, label=left:i] (close-front) ;\nat (4,0) [vowel, label=right:ʉ, label=left:ɨ] (close-central) ;\nat (6,0) [vowel, label=right:u, label=left:ɯ] (close-back) ;\n\nat (-1,-1) Close-mid;\nat (2.5,-1) [vowel, label=right:ø, label=left:e] (cmid-front);\nat (6,-1) [vowel, label=right:o, label=left:ɤ] (cmid-back);\n\nat (-1,-2) Open-mid;\nat (3,-2) [vowel, label=right:œ, label=left:ɛ] (omid-front) ;\nat (6,-2) [vowel, label=right:ɔ, label=left:ʌ] (omid-back) ;\n\nat (-1,-3) Open;\nat (3.5,-3) [vowel, label=left:a] (open-front) ;\nat (6,-3) [vowel, label=right:ɒ, label=left:ɑ] (open-back) ;\n\nat (4.75,-3) (lower-border) ;\nat (4.6,-2.5) ɐ;\nat (4.4,-1.5) ə;\n\n(close-front) – (close-central);\n(close-central) – (close-back);\n(close-back) – (open-back);\n(open-back) – (open-front);\n(open-front) – (close-front) node [very near start, left] æ;\n\n(cmid-front) – (cmid-back);\n(omid-front) – (omid-back);\n\n(close-central) – (lower-border);\n\nCaption: Vowel space\n\nIn this diagram, the two dimensions are height (y-axis) and backness (x-axis), while roundness is determined by the vowel position relative to the reference point: symbols to the left of the dot refer to unrounded vowels and those on the right are rounded.\n\nOther characteristics of vowels\n\nThere are some other features of vowels, usually marked by diacritics or superscript symbols. For example:\n\n- Tongue root position – can have three values:\n\n- neutral = the default tongue root position;\n\n- advanced = when pronouncing the vowel, the tongue root is slightly shifted towards the front. These vowels are called ATR (advanced tongue root) and are marked in the IPA by the diacritic ◌̘;\n\n- retracted = the tongue root is shifted towards the back (retracted). These vowels are called RTR (retracted tongue root) and are marked by ◌̙;"}
{"id":"book_method_c03_s02_p06","kind":"method","chapter":3,"chapter_title":"Phonetics","section":2,"section_title":"Classification of sounds","topic":"phonetics, stress, tone, and versification","text":"- retracted = the tongue root is shifted towards the back (retracted). These vowels are called RTR (retracted tongue root) and are marked by ◌̙;\n\nGenerally, for languages in which the tongue root position is relevant for the vowel articulation, neutral and retracted positions are represented similarly: vowels are treated as either [+ATR] (the tongue root position is advanced) or [−ATR] (the tongue root position is neutral or retracted). Importantly, this feature of vowels is not the same as backness; there can be front [+ATR] vowels, back [+ATR] vowels, front [−ATR] vowels and back [−ATR] vowels. Moreover, this feature is often relevant to languages that display vowel harmony processes (see Section sec:4-vowel-harmony), such as many languages of Africa. In some languages, the following pairs of vowels often pattern as if they were distinguished by the feature [±ATR] (tab:vowelsFeatureATR).\n\nCaption: Vowels and the feature ±ATR\n\n+ATR | −ATR\nu | ʊ\ni | ɪ\no | ɔ\ne | ɛ\na‍ | ə\n\n- Nasalisation is another commonly encountered feature (present, for instance, in French and Portuguese). In the articulation of nasalised vowels, the air escapes through both the mouth and the nose. Nasalised vowels are marked by a tilde () above the vowel symbol: thus, the nasalised variant of the vowel [o] is [õ] (as in the French word bon [bõ]).\n\n- Length can also play an important role in some languages. It refers to the duration of the vowel and we can differentiate at least between short and long vowels (compare Ken and cairn in most non-rhotic English dialects or, for some speakers, the French words mettre and maître).Some languages may have a three-way length distinction. Long vowels are marked in IPA by the symbol [ː] placed after the vowel (this symbol is in fact made up of two triangles pointing towards one another and not by a colon. Nevertheless, in practice, the colon [:] is often used for simplicity). In linguistics problems, a long vowel can also be marked by doubling the vowel (thus, a is the vowel [a] and aa is [aː]) or by a bar above the vowel (ā). The same conventions can apply to long consonants, if they exist.\n\nA final important characteristic of sounds is syllabicity. This denotes the ability of a sound to be the nucleus of a syllable (see below the description of syllable structure). In some languages, vowels are the only possible syllabic sounds. Nevertheless, some consonants can be syllabic in certain languages (e.g., Cantonese) and this feature is marked by a vertical line below the respective consonant (thus, syllabic m is written as [m̩]: consider the English interjection mmmkay for ‘OK’, which we can transcribe as [m̩.keɪ]). We also find syllabic consonants in English words like even [iː.vn̩] or rhythm [ɹɪ.ðm̩]. Conversely, glides such as [w] and [j] are occasionally treated as non-syllabic versions of their respective vowels, which is signalled by an arch diacritic under the symbol: thus, [i̯] and [u̯] are broadly equivalent to [j] and [w].","source_file":"chapters/03-phonetics.tex","source_line_start":13,"license":"CC-BY-4.0","retrieval_text":"Phonetics\nClassification of sounds\nphonetics, stress, tone, and versification\n- retracted = the tongue root is shifted towards the back (retracted). These vowels are called RTR (retracted tongue root) and are marked by ◌̙;\n\nGenerally, for languages in which the tongue root position is relevant for the vowel articulation, neutral and retracted positions are represented similarly: vowels are treated as either [+ATR] (the tongue root position is advanced) or [−ATR] (the tongue root position is neutral or retracted). Importantly, this feature of vowels is not the same as backness; there can be front [+ATR] vowels, back [+ATR] vowels, front [−ATR] vowels and back [−ATR] vowels. Moreover, this feature is often relevant to languages that display vowel harmony processes (see Section sec:4-vowel-harmony), such as many languages of Africa. In some languages, the following pairs of vowels often pattern as if they were distinguished by the feature [±ATR] (tab:vowelsFeatureATR).\n\nCaption: Vowels and the feature ±ATR\n\n+ATR | −ATR\nu | ʊ\ni | ɪ\no | ɔ\ne | ɛ\na‍ | ə\n\n- Nasalisation is another commonly encountered feature (present, for instance, in French and Portuguese). In the articulation of nasalised vowels, the air escapes through both the mouth and the nose. Nasalised vowels are marked by a tilde () above the vowel symbol: thus, the nasalised variant of the vowel [o] is [õ] (as in the French word bon [bõ]).\n\n- Length can also play an important role in some languages. It refers to the duration of the vowel and we can differentiate at least between short and long vowels (compare Ken and cairn in most non-rhotic English dialects or, for some speakers, the French words mettre and maître).Some languages may have a three-way length distinction. Long vowels are marked in IPA by the symbol [ː] placed after the vowel (this symbol is in fact made up of two triangles pointing towards one another and not by a colon. Nevertheless, in practice, the colon [:] is often used for simplicity). In linguistics problems, a long vowel can also be marked by doubling the vowel (thus, a is the vowel [a] and aa is [aː]) or by a bar above the vowel (ā). The same conventions can apply to long consonants, if they exist.\n\nA final important characteristic of sounds is syllabicity. This denotes the ability of a sound to be the nucleus of a syllable (see below the description of syllable structure). In some languages, vowels are the only possible syllabic sounds. Nevertheless, some consonants can be syllabic in certain languages (e.g., Cantonese) and this feature is marked by a vertical line below the respective consonant (thus, syllabic m is written as [m̩]: consider the English interjection mmmkay for ‘OK’, which we can transcribe as [m̩.keɪ]). We also find syllabic consonants in English words like even [iː.vn̩] or rhythm [ɹɪ.ðm̩]. Conversely, glides such as [w] and [j] are occasionally treated as non-syllabic versions of their respective vowels, which is signalled by an arch diacritic under the symbol: thus, [i̯] and [u̯] are broadly equivalent to [j] and [w]."}
{"id":"book_method_c03_s03_p01","kind":"method","chapter":3,"chapter_title":"Phonetics","section":3,"section_title":"Syllable","topic":"phonetics, stress, tone, and versification","text":"A syllable is defined as a group of sounds (phonemes) which are in some way pronounced “together”. In phonetic transcription, the boundary between syllables is marked by a full stop [.]: the word conclusion can be transcribed as [kən.kluː.ʒən]. A syllable is comprised of three parts:\n\n- The nucleus is the “core” of the syllable: the sound occupying the nucleus is always syllabic. (This is generally what is meant when you hear that a syllable can have one and only one vowel: this is true, but as we have seen, a syllabic sound does not have to be a vowel);\n\n- The onset represents the beginning of the syllable (everything before the nucleus);\n\n- The coda represents the end of the syllable (everything after the nucleus).\n\nCaption: Segmentation of the word priest [pri:st].\n\nThe nucleus is the only mandatory component of a syllable: the other two are usually optional. Thus, in English, some syllables consist only of the nucleus (for example the first syllable of upon [ə.pɒn]), some have an onset and a nucleus but no coda (me [mi]), some have a nucleus and a coda but no onset (at [æt]), and yet others have all three (cat [kæt]). A further useful concept is the rhyme, which represents the combination of the nucleus and the coda. For example, a possible segmentation of the word priest [pri:st] is given in Figure fig:priest-syllable.\n\nSyllabification rules\n\nIn this subsection, we will consider the rules that we can follow in placing the syllable boundary. (For simplicity, we will not consider syllabic consonants and will only treat vowels as syllable nuclei.) A syllable generally contains only one vowel (or diphthong):A diphthong is a group consisting of two vowel-like sounds that pattern together as if they were a single vowel, such as in blind [blaɪnd] or lie [laɪ]. therefore, two consecutive vowels will always be part of two different syllables (VV V.V).However, remember that sequences like aa could also represent single instances of long vowels rather than two vowels in a row: check the notes for each problem! If there is a single consonant between two vowels, this will belong to the second syllable (VCV V.CV). This is because the basic principle of syllabification is that a consonant prefers to be an onset rather than a coda.\n\nThings are somewhat more complicated when there are two consonants between a vowel. One possibility is that the syllable boundary goes in the middle of the consonant “cluster” so that one consonant becomes the coda of the first syllable and the other becomes the onset of the second syllable (VCCV VC.CV). In this case, the absence of consonant clusters within the syllable overrides the preference for consonants to be in an onset. Alternatively, the entire cluster can act as an onset, avoiding the coda (VCCV V.CCV). This can also happen, but is usually the more complicated case, in that not all clusters are equally suitable for such a syllabification, so it is perhaps more prudent to assume that VCCV VC.CV is the default pattern.","source_file":"chapters/03-phonetics.tex","source_line_start":261,"license":"CC-BY-4.0","retrieval_text":"Phonetics\nSyllable\nphonetics, stress, tone, and versification\nA syllable is defined as a group of sounds (phonemes) which are in some way pronounced “together”. In phonetic transcription, the boundary between syllables is marked by a full stop [.]: the word conclusion can be transcribed as [kən.kluː.ʒən]. A syllable is comprised of three parts:\n\n- The nucleus is the “core” of the syllable: the sound occupying the nucleus is always syllabic. (This is generally what is meant when you hear that a syllable can have one and only one vowel: this is true, but as we have seen, a syllabic sound does not have to be a vowel);\n\n- The onset represents the beginning of the syllable (everything before the nucleus);\n\n- The coda represents the end of the syllable (everything after the nucleus).\n\nCaption: Segmentation of the word priest [pri:st].\n\nThe nucleus is the only mandatory component of a syllable: the other two are usually optional. Thus, in English, some syllables consist only of the nucleus (for example the first syllable of upon [ə.pɒn]), some have an onset and a nucleus but no coda (me [mi]), some have a nucleus and a coda but no onset (at [æt]), and yet others have all three (cat [kæt]). A further useful concept is the rhyme, which represents the combination of the nucleus and the coda. For example, a possible segmentation of the word priest [pri:st] is given in Figure fig:priest-syllable.\n\nSyllabification rules\n\nIn this subsection, we will consider the rules that we can follow in placing the syllable boundary. (For simplicity, we will not consider syllabic consonants and will only treat vowels as syllable nuclei.) A syllable generally contains only one vowel (or diphthong):A diphthong is a group consisting of two vowel-like sounds that pattern together as if they were a single vowel, such as in blind [blaɪnd] or lie [laɪ]. therefore, two consecutive vowels will always be part of two different syllables (VV V.V).However, remember that sequences like aa could also represent single instances of long vowels rather than two vowels in a row: check the notes for each problem! If there is a single consonant between two vowels, this will belong to the second syllable (VCV V.CV). This is because the basic principle of syllabification is that a consonant prefers to be an onset rather than a coda.\n\nThings are somewhat more complicated when there are two consonants between a vowel. One possibility is that the syllable boundary goes in the middle of the consonant “cluster” so that one consonant becomes the coda of the first syllable and the other becomes the onset of the second syllable (VCCV VC.CV). In this case, the absence of consonant clusters within the syllable overrides the preference for consonants to be in an onset. Alternatively, the entire cluster can act as an onset, avoiding the coda (VCCV V.CCV). This can also happen, but is usually the more complicated case, in that not all clusters are equally suitable for such a syllabification, so it is perhaps more prudent to assume that VCCV VC.CV is the default pattern."}
{"id":"book_method_c03_s04_p01","kind":"method","chapter":3,"chapter_title":"Phonetics","section":4,"section_title":"Versification","topic":"phonetics, stress, tone, and versification","text":"Versification problems are a special type of linguistics problem that involves a set of series of lines (verses) in a particular language, often left even without a translation. Their main purpose is determining the rules governing the structure of the verse (versification, occasionally also prosody, although the latter term has many other meanings). This type of problem can usually be solved using the following method:\n\n- Step 1. Syllabify all the words in each verse. A couple of issues might arise here. First, if you see two vowel symbols next to each other, make sure you know if it is a long vowel, a diphthong (a complex nucleus containing two vowel-like sounds), or a hiatus (a sequence of two vowels belonging to separate syllables). This might be explained in the footnote at the end of the problem. Second, you may need to consider whether the word boundary is relevant: sometimes each word should be syllabified on its own, but sometimes the entire verse should be treated as a single entity.\n\n- Step 2. Determine the type of syllable. With these problems, there are two kinds of criterion that can be used. Sometimes the relevant distinction is between syllables that contain short vowels (V) and those that contain long vowels or a diphthong (VV); sometimes, syllables without a coda contrast with those that have one; sometimes both of these criteria apply. Importantly, the onset is generally irrelevant for this type of problem.\n\n- Step 3. Determine the metre, that is to say, the rules regulating the kinds of syllables a verse can contain, and any restriction on their ordering. Prosodic systems generally use the criteria listed under step 2 to classify syllables as heavy or light syllables, but languages can differ in the details of this process:\n\n- Sometimes, the distinction is solely based on the length of the vowel: light syllables have the structure (C)V(C), while heavy ones have the structure (C)VV(C);\n\n- In other languages, the distinction is solely based on the existence of a coda: a syllable counts as light if it is open, i.e., lacks a coda and thus has the structure (C)V(V), and as heavy if it is closed, having the structure (C)V(V)C;\n\n- In many languages, a syllable counts as heavy if it either contains a long vowel –(C)VV– or if it is closed – (C)V(C)C; in other words if its rhyme contains more than one element;\n\n- Finally, we cannot exclude the possibility that some languages have three types of weight – light syllables (like (C)V), heavy syllables (like (C)VV or (C)VC), and superheavy syllables (like (C)VVC).\n\nAll this information is represented schematically in fig:versification.\n\nCaption: Versification schema.\n\n2-2\nlight | σ\n2-2\n\n2-2\nheavy | lightgrayσ\n2-2\n\n2-2\nsuperheavy | black whiteσ\n2-2\n\n| short | long\nopen | (C)V | lightgray(C)VV\nclosed | (C)VC | lightgray(C)VVC\nType 1\n\n| short | long\nopen | (C)V | (C)VV\nclosed | lightgray(C)VC | lightgray(C)VVC\nType 2\n\n| short | long\nopen | (C)V | lightgray(C)VV\nclosed | lightgray(C)VC | lightgray(C)VVC\nType 3\n\n| short | long\nopen | (C)V | lightgray(C)VV\nclosed | lightgray(C)VC | black white(C)VVC\nType 4\n\nTherefore, we can certainly tell that (C)V syllables will always be light and (C)VVC will always be heavy, while the other two types can belong to either category. The final aim of the problem is to determine the structure of the verse (represented, in general, by a sequence of light and heavy syllables in a particular order). One very important thing to note is that, in many cases, there is an equivalence between one heavy syllable and two light syllables. A good indicator of this phenomenon is if the verses have a variable number of syllables.","source_file":"chapters/03-phonetics.tex","source_line_start":300,"license":"CC-BY-4.0","retrieval_text":"Phonetics\nVersification\nphonetics, stress, tone, and versification\nVersification problems are a special type of linguistics problem that involves a set of series of lines (verses) in a particular language, often left even without a translation. Their main purpose is determining the rules governing the structure of the verse (versification, occasionally also prosody, although the latter term has many other meanings). This type of problem can usually be solved using the following method:\n\n- Step 1. Syllabify all the words in each verse. A couple of issues might arise here. First, if you see two vowel symbols next to each other, make sure you know if it is a long vowel, a diphthong (a complex nucleus containing two vowel-like sounds), or a hiatus (a sequence of two vowels belonging to separate syllables). This might be explained in the footnote at the end of the problem. Second, you may need to consider whether the word boundary is relevant: sometimes each word should be syllabified on its own, but sometimes the entire verse should be treated as a single entity.\n\n- Step 2. Determine the type of syllable. With these problems, there are two kinds of criterion that can be used. Sometimes the relevant distinction is between syllables that contain short vowels (V) and those that contain long vowels or a diphthong (VV); sometimes, syllables without a coda contrast with those that have one; sometimes both of these criteria apply. Importantly, the onset is generally irrelevant for this type of problem.\n\n- Step 3. Determine the metre, that is to say, the rules regulating the kinds of syllables a verse can contain, and any restriction on their ordering. Prosodic systems generally use the criteria listed under step 2 to classify syllables as heavy or light syllables, but languages can differ in the details of this process:\n\n- Sometimes, the distinction is solely based on the length of the vowel: light syllables have the structure (C)V(C), while heavy ones have the structure (C)VV(C);\n\n- In other languages, the distinction is solely based on the existence of a coda: a syllable counts as light if it is open, i.e., lacks a coda and thus has the structure (C)V(V), and as heavy if it is closed, having the structure (C)V(V)C;\n\n- In many languages, a syllable counts as heavy if it either contains a long vowel –(C)VV– or if it is closed – (C)V(C)C; in other words if its rhyme contains more than one element;\n\n- Finally, we cannot exclude the possibility that some languages have three types of weight – light syllables (like (C)V), heavy syllables (like (C)VV or (C)VC), and superheavy syllables (like (C)VVC).\n\nAll this information is represented schematically in fig:versification.\n\nCaption: Versification schema.\n\n2-2\nlight | σ\n2-2\n\n2-2\nheavy | lightgrayσ\n2-2\n\n2-2\nsuperheavy | black whiteσ\n2-2\n\n| short | long\nopen | (C)V | lightgray(C)VV\nclosed | (C)VC | lightgray(C)VVC\nType 1\n\n| short | long\nopen | (C)V | (C)VV\nclosed | lightgray(C)VC | lightgray(C)VVC\nType 2\n\n| short | long\nopen | (C)V | lightgray(C)VV\nclosed | lightgray(C)VC | lightgray(C)VVC\nType 3\n\n| short | long\nopen | (C)V | lightgray(C)VV\nclosed | lightgray(C)VC | black white(C)VVC\nType 4\n\nTherefore, we can certainly tell that (C)V syllables will always be light and (C)VVC will always be heavy, while the other two types can belong to either category. The final aim of the problem is to determine the structure of the verse (represented, in general, by a sequence of light and heavy syllables in a particular order). One very important thing to note is that, in many cases, there is an equivalence between one heavy syllable and two light syllables. A good indicator of this phenomenon is if the verses have a variable number of syllables."}
{"id":"book_method_c03_s05_p01","kind":"method","chapter":3,"chapter_title":"Phonetics","section":5,"section_title":"Stress","topic":"phonetics, stress, tone, and versification","text":"In another type of linguistics problem based solely on phonetic concepts, we are given some words with the stress marked and asked to identify the pattern of stress placement and apply it to other words. Roughly, a stressed syllable is pronounced with more emphasis, allowing us to differentiate between, e.g., contrast (stress on con, as in the noun) and contrast (stress on trast, as in the verb). The stress can also be marked by a prime symbol before the relevant syllable.For example, in the IPA notation, the phonetic transcription of the two words would be [ˈkɒn.trɑːst] and [kən.ˈtrɑːst], respectively. This stress is also called primary stress and, by definition, each word only contains one syllable with a primary stress.\n\nFrom a typological perspective, some languages have fixed stress and others have mobile (or variable) stress. For those which have fixed stress, the stress is usually placed towards the edge of the word (either the beginning or the end). Therefore, there are languages in which the primary stress always falls on the last syllable and languages in which stress always falls on the first syllable. Alternatively, there are languages in which stress always falls on the second or penultimate syllable or even the third or antepenultimate syllable. Generally, the first and last three syllables of a word are the most likely stress positions.\n\nLanguages which have variable stress can be further subdivided into two classes: languages with “free” stress and languages with “predictable” stress. By free stress is meant that on a word-by-word basis you cannot predict the stress pattern (but each word will have a fixed pattern). In languages with “predictable” stress, the stress placement follows some well-defined rules, usually related to the weight of syllables. In linguistics problems, we usually find languages with predictable stress, because free stress cannot be predicted, while fixed stress is extremely easy to analyse.\n\nIn the case of languages with predictable stress, stress is usually placed within a window (i.e., a group of syllables, with the accent always falling on one of these syllables, based on certain rules). The most common stress windows are the first and the last three syllables and, within these windows, certain rules dictate stress placement, which are often similar to the rules in versification problems. For instance, stress could always fall on the syllable that contains a long vowel (within a certain stress window), and, if there is no such syllable, it will be placed on the first syllable within that window. Generally, all these rules are based on syllable weight (i.e., vowel length and/or the presence of a coda). It is possible, but rare, for stress placement to also be determined by factors such as the quality of the vowel (front/back, open/close, etc.).","source_file":"chapters/03-phonetics.tex","source_line_start":757,"license":"CC-BY-4.0","retrieval_text":"Phonetics\nStress\nphonetics, stress, tone, and versification\nIn another type of linguistics problem based solely on phonetic concepts, we are given some words with the stress marked and asked to identify the pattern of stress placement and apply it to other words. Roughly, a stressed syllable is pronounced with more emphasis, allowing us to differentiate between, e.g., contrast (stress on con, as in the noun) and contrast (stress on trast, as in the verb). The stress can also be marked by a prime symbol before the relevant syllable.For example, in the IPA notation, the phonetic transcription of the two words would be [ˈkɒn.trɑːst] and [kən.ˈtrɑːst], respectively. This stress is also called primary stress and, by definition, each word only contains one syllable with a primary stress.\n\nFrom a typological perspective, some languages have fixed stress and others have mobile (or variable) stress. For those which have fixed stress, the stress is usually placed towards the edge of the word (either the beginning or the end). Therefore, there are languages in which the primary stress always falls on the last syllable and languages in which stress always falls on the first syllable. Alternatively, there are languages in which stress always falls on the second or penultimate syllable or even the third or antepenultimate syllable. Generally, the first and last three syllables of a word are the most likely stress positions.\n\nLanguages which have variable stress can be further subdivided into two classes: languages with “free” stress and languages with “predictable” stress. By free stress is meant that on a word-by-word basis you cannot predict the stress pattern (but each word will have a fixed pattern). In languages with “predictable” stress, the stress placement follows some well-defined rules, usually related to the weight of syllables. In linguistics problems, we usually find languages with predictable stress, because free stress cannot be predicted, while fixed stress is extremely easy to analyse.\n\nIn the case of languages with predictable stress, stress is usually placed within a window (i.e., a group of syllables, with the accent always falling on one of these syllables, based on certain rules). The most common stress windows are the first and the last three syllables and, within these windows, certain rules dictate stress placement, which are often similar to the rules in versification problems. For instance, stress could always fall on the syllable that contains a long vowel (within a certain stress window), and, if there is no such syllable, it will be placed on the first syllable within that window. Generally, all these rules are based on syllable weight (i.e., vowel length and/or the presence of a coda). It is possible, but rare, for stress placement to also be determined by factors such as the quality of the vowel (front/back, open/close, etc.)."}
{"id":"book_method_c03_s05_p02","kind":"method","chapter":3,"chapter_title":"Phonetics","section":5,"section_title":"Stress","topic":"phonetics, stress, tone, and versification","text":"In the case of languages with predictable stress, stress is usually placed within a window (i.e., a group of syllables, with the accent always falling on one of these syllables, based on certain rules). The most common stress windows are the first and the last three syllables and, within these windows, certain rules dictate stress placement, which are often similar to the rules in versification problems. For instance, stress could always fall on the syllable that contains a long vowel (within a certain stress window), and, if there is no such syllable, it will be placed on the first syllable within that window. Generally, all these rules are based on syllable weight (i.e., vowel length and/or the presence of a coda). It is possible, but rare, for stress placement to also be determined by factors such as the quality of the vowel (front/back, open/close, etc.).\n\nTo give you an idea of how complex such systems can be, let us consider the case of the Pirahã language.This phenomenon was featured in a problem by Artūrs Semeņuks (IOL 2013). In this language, the stress window is represented by the final three syllables (thus, the primary stress will always be placed on one of the last three syllables). The placement of the stress is determined by a hierarchy of syllable types that takes into account both the type of vowel (long or short) and the type of onset (an extremely rare thing). Specifically, syllables with a long vowel are always higher in the hierarchy (``heavier\") than those with a short vowel, while syllables that feature a voiceless onset are higher than those that have a voiced onset (which are themselves higher than those which have no onset). If we denote a voiced consonant as G and a voiceless consonant as C (and, as previously, use V for a short vowel and VV for a long vowel or diphthong), the syllable hierarchy in Pirahã is: CVV > GVV > VV > CV > GV > V. Furthermore, if there are several syllables of the same type within the window, stress will be placed on the rightmost syllable (last > penultimate > antepenultimate). For example, in the word [ka.gi.hi] ‘wasp’ we have two CV syllables ([ka] and [hi]) and one GV syllable ([gi]]. We know that CV syllables have priority, so, certainly, the stress will not be placed on [gi]. Amongst the remaining two syllables (which are of the same type), the stress will be placed on the rightmost syllable, so the word is stressed [ka.gi.ˈhi].\n\nBy contrast in [ʔi.soo.bai] ‘otter’ stress goes on [soo]: it has priority over the first syllable since it has a long vowel and it also has priority over the last syllable since it has a voiceless onset.\n\nApart from primary stress, some languages can also have secondary stresses. These are also marked with a prime symbol, but placed below the line (ˌ). There is no limit as to how many secondary stresses a word can have and usually the rules for placing the secondary stress are based on the type of syllable or vowel (e.g., all syllables with a long vowel that do not bear primary stress will receive secondary stress) or related to the primary stress (e.g., starting from the primary stress towards the right, every other syllable receives secondary stress).","source_file":"chapters/03-phonetics.tex","source_line_start":757,"license":"CC-BY-4.0","retrieval_text":"Phonetics\nStress\nphonetics, stress, tone, and versification\nIn the case of languages with predictable stress, stress is usually placed within a window (i.e., a group of syllables, with the accent always falling on one of these syllables, based on certain rules). The most common stress windows are the first and the last three syllables and, within these windows, certain rules dictate stress placement, which are often similar to the rules in versification problems. For instance, stress could always fall on the syllable that contains a long vowel (within a certain stress window), and, if there is no such syllable, it will be placed on the first syllable within that window. Generally, all these rules are based on syllable weight (i.e., vowel length and/or the presence of a coda). It is possible, but rare, for stress placement to also be determined by factors such as the quality of the vowel (front/back, open/close, etc.).\n\nTo give you an idea of how complex such systems can be, let us consider the case of the Pirahã language.This phenomenon was featured in a problem by Artūrs Semeņuks (IOL 2013). In this language, the stress window is represented by the final three syllables (thus, the primary stress will always be placed on one of the last three syllables). The placement of the stress is determined by a hierarchy of syllable types that takes into account both the type of vowel (long or short) and the type of onset (an extremely rare thing). Specifically, syllables with a long vowel are always higher in the hierarchy (``heavier\") than those with a short vowel, while syllables that feature a voiceless onset are higher than those that have a voiced onset (which are themselves higher than those which have no onset). If we denote a voiced consonant as G and a voiceless consonant as C (and, as previously, use V for a short vowel and VV for a long vowel or diphthong), the syllable hierarchy in Pirahã is: CVV > GVV > VV > CV > GV > V. Furthermore, if there are several syllables of the same type within the window, stress will be placed on the rightmost syllable (last > penultimate > antepenultimate). For example, in the word [ka.gi.hi] ‘wasp’ we have two CV syllables ([ka] and [hi]) and one GV syllable ([gi]]. We know that CV syllables have priority, so, certainly, the stress will not be placed on [gi]. Amongst the remaining two syllables (which are of the same type), the stress will be placed on the rightmost syllable, so the word is stressed [ka.gi.ˈhi].\n\nBy contrast in [ʔi.soo.bai] ‘otter’ stress goes on [soo]: it has priority over the first syllable since it has a long vowel and it also has priority over the last syllable since it has a voiceless onset.\n\nApart from primary stress, some languages can also have secondary stresses. These are also marked with a prime symbol, but placed below the line (ˌ). There is no limit as to how many secondary stresses a word can have and usually the rules for placing the secondary stress are based on the type of syllable or vowel (e.g., all syllables with a long vowel that do not bear primary stress will receive secondary stress) or related to the primary stress (e.g., starting from the primary stress towards the right, every other syllable receives secondary stress)."}
{"id":"book_method_c03_s06_p01","kind":"method","chapter":3,"chapter_title":"Phonetics","section":6,"section_title":"Tone","topic":"phonetics, stress, tone, and versification","text":"Another phonetic phenomenon is tone. This term covers the use of pitch to make distinctions in the meaning of words, and/or grammatical distinctions.\n\nTechnically, pitch is the frequency with which the vocal folds vibrate in the production of speech, also known as the fundamental frequency (for this reason, only voiced sounds are characterised by pitch). Most, if not all, languages make some use of pitch. For example, in many languages syllables that carry stress are characterised by a higher pitch than unstressed ones: most varieties of English belong to this type. In some languages, such as for example Russian, the pitch trajectory is the only difference between a statement and a general question, in contrast to, say, English, which uses a special construction to distinguish between You solved the problem and Did you solve the problem?\n\nSuch uses of pitch are found in many languages of the world, and are covered by the label intonation. The term tone is narrower: we say that a language is a tone language if tone can make lexical or grammatical distinctions between words. For example, the syllable ma in Mandarin can have different meanings depending on the tone it receives (as we have seen in the previous chapter). As such, the word mā, with first tone, means ‘mother’, while the word má (second tone) means ‘hemp’, mǎ (third tone) means ‘horse’, and mà (fourth tone) means ‘scold’.\n\nIn such cases, tonal differences are not really different from distinctions made using, say, two different consonants. There are also languages where changes in tone can express grammatical meanings, similar to how such meaning can be expressed by a morpheme (see chap-noun for more information about morphemes).\n\nThe terminology used for tone languages can differ quite a lot, so you may need to read the pronunciation notes carefully to make sure you understand the notation in the problem.\n\nTone, like stress, is an example of a suprasegmental phenomenon. Whilst properties such as, say, backness or place of articulation belong only to the vowel or consonant they are associated with, suprasegmental features characterise groups of consonants and vowels (collectively called segments). With stress, we saw that it is mostly a property of syllables, and this is often the case also with tone.","source_file":"chapters/03-phonetics.tex","source_line_start":888,"license":"CC-BY-4.0","retrieval_text":"Phonetics\nTone\nphonetics, stress, tone, and versification\nAnother phonetic phenomenon is tone. This term covers the use of pitch to make distinctions in the meaning of words, and/or grammatical distinctions.\n\nTechnically, pitch is the frequency with which the vocal folds vibrate in the production of speech, also known as the fundamental frequency (for this reason, only voiced sounds are characterised by pitch). Most, if not all, languages make some use of pitch. For example, in many languages syllables that carry stress are characterised by a higher pitch than unstressed ones: most varieties of English belong to this type. In some languages, such as for example Russian, the pitch trajectory is the only difference between a statement and a general question, in contrast to, say, English, which uses a special construction to distinguish between You solved the problem and Did you solve the problem?\n\nSuch uses of pitch are found in many languages of the world, and are covered by the label intonation. The term tone is narrower: we say that a language is a tone language if tone can make lexical or grammatical distinctions between words. For example, the syllable ma in Mandarin can have different meanings depending on the tone it receives (as we have seen in the previous chapter). As such, the word mā, with first tone, means ‘mother’, while the word má (second tone) means ‘hemp’, mǎ (third tone) means ‘horse’, and mà (fourth tone) means ‘scold’.\n\nIn such cases, tonal differences are not really different from distinctions made using, say, two different consonants. There are also languages where changes in tone can express grammatical meanings, similar to how such meaning can be expressed by a morpheme (see chap-noun for more information about morphemes).\n\nThe terminology used for tone languages can differ quite a lot, so you may need to read the pronunciation notes carefully to make sure you understand the notation in the problem.\n\nTone, like stress, is an example of a suprasegmental phenomenon. Whilst properties such as, say, backness or place of articulation belong only to the vowel or consonant they are associated with, suprasegmental features characterise groups of consonants and vowels (collectively called segments). With stress, we saw that it is mostly a property of syllables, and this is often the case also with tone."}
{"id":"book_method_c04_s01_p01","kind":"method","chapter":4,"chapter_title":"Phonology","section":1,"section_title":"Introduction","topic":"phonological rules and sound correspondences","text":"Whilst phonetics, the subject of the previous chapter, is concerned with the physical aspects of sound, phonology is the study of their functions and the ways in which sounds can pattern within the system of the language. Phonology is connected to phonetics in the sense that, once the basic phonetic concepts are understood (characterisation of the sounds), we can begin unveiling the ways in which these sounds interact with one another. It will therefore be useful to introduce a way to succinctly describe the phonological changes that can take place.","source_file":"chapters/04-phonology.tex","source_line_start":6,"license":"CC-BY-4.0","retrieval_text":"Phonology\nIntroduction\nphonological rules and sound correspondences\nWhilst phonetics, the subject of the previous chapter, is concerned with the physical aspects of sound, phonology is the study of their functions and the ways in which sounds can pattern within the system of the language. Phonology is connected to phonetics in the sense that, once the basic phonetic concepts are understood (characterisation of the sounds), we can begin unveiling the ways in which these sounds interact with one another. It will therefore be useful to introduce a way to succinctly describe the phonological changes that can take place."}
{"id":"book_method_c04_s02_p01","kind":"method","chapter":4,"chapter_title":"Phonology","section":2,"section_title":"Phonological notation","topic":"phonological rules and sound correspondences","text":"NotebulbonThese notations are meant to make rule writing easier and faster, thus saving time. However, their use is not necessary; if you are unsure of how to use them, it is better to avoid using them in an actual competition, rather than risk using them incorrectly.\n\n±A\nThis shows that the sound we refer to has (+) or does not have (–) the\nfeature A. Thus, we can characterise the sound [m] (bilabial\nnasal) as\n+bilabial\n+nasal,\n\nand the sound [s] (voiceless alveolar fricative) as\n+alveolar\n–voice\n+fricative.\n\nNotebulbonIn reality, the notation is more complex in the sense\nthat each feature A must be binary (i.e., have only two\npossible values). A feature like voice is binary in that all consonants are either voiced (+voice)\nor voiceless (-voice). The parameter fricative is somewhat different: the sounds that are not fricatives (-fricative) do not necessarily share a common property, e.g., they can be stops, affricates, nasals, and so on. Nevertheless, for the purpose of linguistics problems, there is no need to get into any more details and we can simply write the place and manner of articulation.\n\nA B / C _ V This is a phonological rule and is to be read as follows: ``the sound A becomes () B if it appears in a certain environment (/)...\" (in this\ncase, if it is between a consonant (C) and a vowel (V)). When writing phonological rules, the underscore ( _ ) shows the position of the sound which is affected. Therefore, if we want to write the rule\n``k becomes g before a'', we write k g / _ a.\n\nThere are other symbols that can be used to explain the environment in which the transformation takes place. For example, the hash sign (#) marks the word boundary (so _ # represents the end of the word – the sound that changes is before the word boundary, so at the end of the word –, while # _ represents the beginning of the word). As mentioned in the previous chapter, the syllable boundary is marked with a full stop (.). The notation C0 represents a sequence of consonants (or an absence of a consonant): thus, the string uC0 matches u, uC, uCC, uCCC, etc.\n\nIn order to denote multiple possibilities, we can use curly brackets. If we want to write the rule ``k becomes g if at the beginning of the word or after a vowel'', we can write:\n\nkg{#\nV\n}\n\nWhen writing these rules, we can also use the (square) bracket notation shown above, therefore, if we want to write ``k becomes g after a voiced consonant'', we can write:\n\nkg–syllabic\n+voice\n\nWe need to mention the parameter [-syllabic] (specifying that it is a non-syllabic sound) in order to exclude the vowels.\n\nMoreover, in certain cases, we can use Greek letters\n(α, , , etc.) instead of . They are used in\norder to show that a certain feature has the same value in multiple parts of the rule, but it could be either a plus or a minus but, importantly, the values must match.\n\nOne common use of this notation is to describe assimilation. For example, the following rule:\n\nnα PLACE+stop\nα PLACE\ncan be read as: ``The consonant n before a stop will change its place of articulation to match the place of articulation of the stop.\" Basically, this rule combines all the following into one:\n\n- n+bilabial+stop\n+bilabial\n\n- n+palatal+stop\n+palatal\n\n- n+velar+stop\n+velar\nand so on.\n\nFor more examples of how to use alpha notation, see problems 4.3 and 4.4.","source_file":"chapters/04-phonology.tex","source_line_start":11,"license":"CC-BY-4.0","retrieval_text":"Phonology\nPhonological notation\nphonological rules and sound correspondences\nNotebulbonThese notations are meant to make rule writing easier and faster, thus saving time. However, their use is not necessary; if you are unsure of how to use them, it is better to avoid using them in an actual competition, rather than risk using them incorrectly.\n\n±A\nThis shows that the sound we refer to has (+) or does not have (–) the\nfeature A. Thus, we can characterise the sound [m] (bilabial\nnasal) as\n+bilabial\n+nasal,\n\nand the sound [s] (voiceless alveolar fricative) as\n+alveolar\n–voice\n+fricative.\n\nNotebulbonIn reality, the notation is more complex in the sense\nthat each feature A must be binary (i.e., have only two\npossible values). A feature like voice is binary in that all consonants are either voiced (+voice)\nor voiceless (-voice). The parameter fricative is somewhat different: the sounds that are not fricatives (-fricative) do not necessarily share a common property, e.g., they can be stops, affricates, nasals, and so on. Nevertheless, for the purpose of linguistics problems, there is no need to get into any more details and we can simply write the place and manner of articulation.\n\nA B / C _ V This is a phonological rule and is to be read as follows: ``the sound A becomes () B if it appears in a certain environment (/)...\" (in this\ncase, if it is between a consonant (C) and a vowel (V)). When writing phonological rules, the underscore ( _ ) shows the position of the sound which is affected. Therefore, if we want to write the rule\n``k becomes g before a'', we write k g / _ a.\n\nThere are other symbols that can be used to explain the environment in which the transformation takes place. For example, the hash sign (#) marks the word boundary (so _ # represents the end of the word – the sound that changes is before the word boundary, so at the end of the word –, while # _ represents the beginning of the word). As mentioned in the previous chapter, the syllable boundary is marked with a full stop (.). The notation C0 represents a sequence of consonants (or an absence of a consonant): thus, the string uC0 matches u, uC, uCC, uCCC, etc.\n\nIn order to denote multiple possibilities, we can use curly brackets. If we want to write the rule ``k becomes g if at the beginning of the word or after a vowel'', we can write:\n\nkg{#\nV\n}\n\nWhen writing these rules, we can also use the (square) bracket notation shown above, therefore, if we want to write ``k becomes g after a voiced consonant'', we can write:\n\nkg–syllabic\n+voice\n\nWe need to mention the parameter [-syllabic] (specifying that it is a non-syllabic sound) in order to exclude the vowels.\n\nMoreover, in certain cases, we can use Greek letters\n(α, , , etc.) instead of . They are used in\norder to show that a certain feature has the same value in multiple parts of the rule, but it could be either a plus or a minus but, importantly, the values must match.\n\nOne common use of this notation is to describe assimilation. For example, the following rule:\n\nnα PLACE+stop\nα PLACE\ncan be read as: ``The consonant n before a stop will change its place of articulation to match the place of articulation of the stop.\" Basically, this rule combines all the following into one:\n\n- n+bilabial+stop\n+bilabial\n\n- n+palatal+stop\n+palatal\n\n- n+velar+stop\n+velar\nand so on.\n\nFor more examples of how to use alpha notation, see problems 4.3 and 4.4."}
{"id":"book_method_c04_s03_p01","kind":"method","chapter":4,"chapter_title":"Phonology","section":3,"section_title":"Complementary distributionComplementary distribution","topic":"phonological rules and sound correspondences","text":"Complementary distribution is the relation between two or more sounds, in which each sound can only be found in certain environments (under certain conditions). Some languages can use the same symbol for two or more sounds which have a complementary distribution because, depending on the environment where they are found, there will be no ambiguity regarding the sound it represents. For example, in Korean, the character ㄹ represents both the sounds l and r, but there are specific pronunciation rules. This character is pronounced r if it is between two vowels, and l otherwise. As one can notice, there is a predictable environment (which can be described) and another default environment (for all the other cases). We can write a phonological rule to explain the transformation of the sound in the given context. For example, for Korean, we can write the rule: l r / V _ V (reading “the sound l becomes r if it is between two vowels”). In all cases, the transformation is made starting from the default sound (whose environment is not predictable) towards the predictable sound (so we can then specify the exact environment in which the transformation takes place).\n\nOften, complementary distribution environments depend on the position before/after a vowel or the beginning/end of the word. Thus, these are the first things to check. Of course, there can be more complex cases in which the complementary distribution depends on the features of the previous or following sounds or even sounds further away in the word.\n\nLet us consider the following examples from Spanish. We focus on the sounds d and ð, which, in these data, are in a complementary distribution:\n\naban[d]onar, alcal[d]e, [d]ecir, [d]oncel, [d]on[d]e, entra[ð]a, la[ð]o, me[ð]ir, na[ð]a, senti[ð]o\n\nThe first step is to split the words into two groups, depending on which sound they contain:\n\nd sound | ð sound\naban[d]onar | entra[ð]a\nalcal[d]e | la[ð]o\n[d]ecir | me[ð]ir\n[d]oncel | na[ð]a\n[d]on[d]e | senti[ð]o\n\nWe notice that neither of the two sounds appears exclusively at the end or beginning of the word, so we can exclude the possibility that this is the relevant factor. Next, we check if any of the two prefer to be before/after/in between vowels. We notice that both d and ð appear before vowels, so this is not helpful, but we notice that ð is found only after a vowel (while d is always either at the beginning of the word, or after a consonant). Thus, since ð is the one with a predictable environment (after a vowel), we can write the phonological rule: d ð / V_.\n\nLet us analyse the following examples (for simplicity, they are already split into two columns):\n\nIn Kimatuumbi (spoken in Tanzania), sounds g and ɠ are\nin a complementary distribution.\n\ng sound | ɠ sound\nlisɛɛŋgɛlɛ | ɠʊlʊja\nkjaaŋgi | nuɠa\nlikʊʊŋgwa | ɠɔlɔja\nŋgaambalɛ | ɠʊlʊka\n\nIn this case, we can easily observe that g only appears after\nŋ. Thus, we can write the rule: ɠ g / ŋ _.\n\nIn a variety of Luganda, the sounds l and r are in a complementary distribution:\n\nl sound | r sound\nkola | beera\nlwana | jjukira\nlja | erjato\nluula | effirimbi\nomugole | emmeeri\nlumonde | eraddu\noluganda | wawaabira","source_file":"chapters/04-phonology.tex","source_line_start":113,"license":"CC-BY-4.0","retrieval_text":"Phonology\nComplementary distributionComplementary distribution\nphonological rules and sound correspondences\nComplementary distribution is the relation between two or more sounds, in which each sound can only be found in certain environments (under certain conditions). Some languages can use the same symbol for two or more sounds which have a complementary distribution because, depending on the environment where they are found, there will be no ambiguity regarding the sound it represents. For example, in Korean, the character ㄹ represents both the sounds l and r, but there are specific pronunciation rules. This character is pronounced r if it is between two vowels, and l otherwise. As one can notice, there is a predictable environment (which can be described) and another default environment (for all the other cases). We can write a phonological rule to explain the transformation of the sound in the given context. For example, for Korean, we can write the rule: l r / V _ V (reading “the sound l becomes r if it is between two vowels”). In all cases, the transformation is made starting from the default sound (whose environment is not predictable) towards the predictable sound (so we can then specify the exact environment in which the transformation takes place).\n\nOften, complementary distribution environments depend on the position before/after a vowel or the beginning/end of the word. Thus, these are the first things to check. Of course, there can be more complex cases in which the complementary distribution depends on the features of the previous or following sounds or even sounds further away in the word.\n\nLet us consider the following examples from Spanish. We focus on the sounds d and ð, which, in these data, are in a complementary distribution:\n\naban[d]onar, alcal[d]e, [d]ecir, [d]oncel, [d]on[d]e, entra[ð]a, la[ð]o, me[ð]ir, na[ð]a, senti[ð]o\n\nThe first step is to split the words into two groups, depending on which sound they contain:\n\nd sound | ð sound\naban[d]onar | entra[ð]a\nalcal[d]e | la[ð]o\n[d]ecir | me[ð]ir\n[d]oncel | na[ð]a\n[d]on[d]e | senti[ð]o\n\nWe notice that neither of the two sounds appears exclusively at the end or beginning of the word, so we can exclude the possibility that this is the relevant factor. Next, we check if any of the two prefer to be before/after/in between vowels. We notice that both d and ð appear before vowels, so this is not helpful, but we notice that ð is found only after a vowel (while d is always either at the beginning of the word, or after a consonant). Thus, since ð is the one with a predictable environment (after a vowel), we can write the phonological rule: d ð / V_.\n\nLet us analyse the following examples (for simplicity, they are already split into two columns):\n\nIn Kimatuumbi (spoken in Tanzania), sounds g and ɠ are\nin a complementary distribution.\n\ng sound | ɠ sound\nlisɛɛŋgɛlɛ | ɠʊlʊja\nkjaaŋgi | nuɠa\nlikʊʊŋgwa | ɠɔlɔja\nŋgaambalɛ | ɠʊlʊka\n\nIn this case, we can easily observe that g only appears after\nŋ. Thus, we can write the rule: ɠ g / ŋ _.\n\nIn a variety of Luganda, the sounds l and r are in a complementary distribution:\n\nl sound | r sound\nkola | beera\nlwana | jjukira\nlja | erjato\nluula | effirimbi\nomugole | emmeeri\nlumonde | eraddu\noluganda | wawaabira"}
{"id":"book_method_c04_s03_p02","kind":"method","chapter":4,"chapter_title":"Phonology","section":3,"section_title":"Complementary distributionComplementary distribution","topic":"phonological rules and sound correspondences","text":"l sound | r sound\nkola | beera\nlwana | jjukira\nlja | erjato\nluula | effirimbi\nomugole | emmeeri\nlumonde | eraddu\noluganda | wawaabira\n\nThis time, we notice that both l and r appear solely next to vowels (or j). Moreover, their environment does not seem to be connected to the beginning or end of the word, so the next step is to look at the type of sounds next to which they appear. We notice that l appears only at the beginning of the word or after the vowel o, while r appears only after the vowels i and e. Therefore, based on this observation, we can write two possible phonological rules: l r / {e, i} _ or r l / {# _, o _}. This is the moment when it is important to understand the phonetic concepts. Generally, phonological rules are conditioned by a particular feature or features of sounds in the environment where the alternating sounds appear. Thus, we notice that both i and e are front vowels (and the data do not contain any other front vowels), so the rule becomes: l r / [+ front] _. Generally, if an environment depends on more sounds (in this case i and e), there will be a connection between these sounds (they will share a certain feature) and, in most cases, in linguistics problems, there will be a task asking you to extend this rule to other sounds. For example, for this problem, it is possible that one of the tasks would feature another front vowel, and you will need to deduce that after that vowel the sound r will be used.","source_file":"chapters/04-phonology.tex","source_line_start":113,"license":"CC-BY-4.0","retrieval_text":"Phonology\nComplementary distributionComplementary distribution\nphonological rules and sound correspondences\nl sound | r sound\nkola | beera\nlwana | jjukira\nlja | erjato\nluula | effirimbi\nomugole | emmeeri\nlumonde | eraddu\noluganda | wawaabira\n\nThis time, we notice that both l and r appear solely next to vowels (or j). Moreover, their environment does not seem to be connected to the beginning or end of the word, so the next step is to look at the type of sounds next to which they appear. We notice that l appears only at the beginning of the word or after the vowel o, while r appears only after the vowels i and e. Therefore, based on this observation, we can write two possible phonological rules: l r / {e, i} _ or r l / {# _, o _}. This is the moment when it is important to understand the phonetic concepts. Generally, phonological rules are conditioned by a particular feature or features of sounds in the environment where the alternating sounds appear. Thus, we notice that both i and e are front vowels (and the data do not contain any other front vowels), so the rule becomes: l r / [+ front] _. Generally, if an environment depends on more sounds (in this case i and e), there will be a connection between these sounds (they will share a certain feature) and, in most cases, in linguistics problems, there will be a task asking you to extend this rule to other sounds. For example, for this problem, it is possible that one of the tasks would feature another front vowel, and you will need to deduce that after that vowel the sound r will be used."}
{"id":"book_method_c04_s04_p01","kind":"method","chapter":4,"chapter_title":"Phonology","section":4,"section_title":"Phonological\nprocessesPhonological processes","topic":"phonological rules and sound correspondences","text":"Below we will present and discuss some of the most common phonological changes. These can be classified into four groups, depending on the effect they have:\n\n- Deletion of a sound. When the deleted sound is a vowel, you might encounter some more specific terms, such as:\n\n- Aphaeresis: deletion at the beginning of the word;\n\n- Syncope: deletion in the middle of the word, especially when another syllable follows the deleted vowel;\n\n- Apocope: deletion at the end of the word.\n\nIn linguistics problems, we can use, for simplicity, the term deletion, as long as we specify where it takes place and in which environment. One common kind of deletion occurs when two identical sounds come in contact with one another. For example, in Ainu, the suffix -re becomes -e if it is added to a word that already ends in -r. Thus, an r is deleted in order to avoid two consecutive identical sounds. We can write: r / r _ e.\n\n- Insertion of a sound, also known as epenthesis.\n\nIn linguistics problems, epenthesis often occurs between two consonants or two vowels, as well as word-initially (this particular type is sometimes called prosthesis). Thus, if a stem that ends in a consonant adds a suffix that starts with a consonant, it is possible to add an epenthetic vowel (and avoid two consecutive consonants). For example, most English nouns form the plural with the addition of a single consonant (cats, dogs), but in words like horses there is an extra vowel before the plural marker.\n\nThe same thing may happen between two vowels, as in British English varieties that have so-called linking and intrusive r (an example of intrusive r is drawing draw-r-ing).\n\n- Metathesis = process that causes the transposition of two or more sounds.\n\nFor example, the word third suffered a metathesis from Old English (thridda), in which the sounds i and r switched places. Although in general, the sounds that switch places are next to one another, there are also cases when metathesis can occur at a considerable distance: for instance, in the Tertenia dialect of Sardinian the word ‘belly’ occurs as brεnti in isolation but as (b)εntri after the definite article sa.\n\nAnother example is the Romanian word întreg (‘whole’). It comes from the Latin word integrum in which the sound r and the sequence eg switched places.\n\n- The last and most important category is that of processes in which one sound is transformed because of another sound in its vicinity. These processes are generally called assimilations. Assimilations can be classified based on different criteria as follows:\n\n- assimilation can affect both vowels and consonants;\n\n- based on the degree of assimilation we can have total assimilation (in which the target sound becomes completely identical to the trigger) or partial assimilation (in which only certain features are changed);\n\n- based on the position of the sound: contact assimilation (the two sounds are next to one another) or assimilation at a distance. Assimilation at a distance is often referred to as harmony; vowel harmony is quite common, while consonant harmony exists, but is relatively rare;\n\n- based on the direction of the assimilation: progressive (in which the sound after is changed as a result of its interaction with a previous sound) or regressive (in which the sound that changes occurs before the sound that triggers the assimilation).\n\nOne of the most common examples of assimilation is the partial assimilation of nasals in consonant clusters, as exemplified before. Nasal consonants are extremely prone to assimilation, and they often assimilate to the place of articulation of the neighbouring consonant. This phenomenon also occurs in English where, for example, can be is pronounced [kæmbi] in fast speech, in which case the nasal n “borrows” the place of articulation from the following consonant (b, bilabial) and becomes m.","source_file":"chapters/04-phonology.tex","source_line_start":446,"license":"CC-BY-4.0","retrieval_text":"Phonology\nPhonological\nprocessesPhonological processes\nphonological rules and sound correspondences\nBelow we will present and discuss some of the most common phonological changes. These can be classified into four groups, depending on the effect they have:\n\n- Deletion of a sound. When the deleted sound is a vowel, you might encounter some more specific terms, such as:\n\n- Aphaeresis: deletion at the beginning of the word;\n\n- Syncope: deletion in the middle of the word, especially when another syllable follows the deleted vowel;\n\n- Apocope: deletion at the end of the word.\n\nIn linguistics problems, we can use, for simplicity, the term deletion, as long as we specify where it takes place and in which environment. One common kind of deletion occurs when two identical sounds come in contact with one another. For example, in Ainu, the suffix -re becomes -e if it is added to a word that already ends in -r. Thus, an r is deleted in order to avoid two consecutive identical sounds. We can write: r / r _ e.\n\n- Insertion of a sound, also known as epenthesis.\n\nIn linguistics problems, epenthesis often occurs between two consonants or two vowels, as well as word-initially (this particular type is sometimes called prosthesis). Thus, if a stem that ends in a consonant adds a suffix that starts with a consonant, it is possible to add an epenthetic vowel (and avoid two consecutive consonants). For example, most English nouns form the plural with the addition of a single consonant (cats, dogs), but in words like horses there is an extra vowel before the plural marker.\n\nThe same thing may happen between two vowels, as in British English varieties that have so-called linking and intrusive r (an example of intrusive r is drawing draw-r-ing).\n\n- Metathesis = process that causes the transposition of two or more sounds.\n\nFor example, the word third suffered a metathesis from Old English (thridda), in which the sounds i and r switched places. Although in general, the sounds that switch places are next to one another, there are also cases when metathesis can occur at a considerable distance: for instance, in the Tertenia dialect of Sardinian the word ‘belly’ occurs as brεnti in isolation but as (b)εntri after the definite article sa.\n\nAnother example is the Romanian word întreg (‘whole’). It comes from the Latin word integrum in which the sound r and the sequence eg switched places.\n\n- The last and most important category is that of processes in which one sound is transformed because of another sound in its vicinity. These processes are generally called assimilations. Assimilations can be classified based on different criteria as follows:\n\n- assimilation can affect both vowels and consonants;\n\n- based on the degree of assimilation we can have total assimilation (in which the target sound becomes completely identical to the trigger) or partial assimilation (in which only certain features are changed);\n\n- based on the position of the sound: contact assimilation (the two sounds are next to one another) or assimilation at a distance. Assimilation at a distance is often referred to as harmony; vowel harmony is quite common, while consonant harmony exists, but is relatively rare;\n\n- based on the direction of the assimilation: progressive (in which the sound after is changed as a result of its interaction with a previous sound) or regressive (in which the sound that changes occurs before the sound that triggers the assimilation).\n\nOne of the most common examples of assimilation is the partial assimilation of nasals in consonant clusters, as exemplified before. Nasal consonants are extremely prone to assimilation, and they often assimilate to the place of articulation of the neighbouring consonant. This phenomenon also occurs in English where, for example, can be is pronounced [kæmbi] in fast speech, in which case the nasal n “borrows” the place of articulation from the following consonant (b, bilabial) and becomes m."}
{"id":"book_method_c04_s05_p01","kind":"method","chapter":4,"chapter_title":"Phonology","section":5,"section_title":"Vowel harmony","topic":"phonological rules and sound correspondences","text":"It is interesting to consider the reason why the stem changes occur. These changes only happen for non-future and imperative, and these are the only forms for which the corresponding suffixes begin with a consonant. Thus, the epenthesis (rule 1b) is rather common, and it is used to avoid a three-consonant cluster, since this would impede the pronunciation.\n\nThe vowel shortening, on the other hand, highlights a much more interesting phenomenon which can be connected to syllable weight. Remember that in the previous chapter, we classified syllables into four types, depending on vowel length and on the existence of a coda. In this language, it would appear that syllables which have both a long vowel and a consonant are not allowed; moreover, the onset and the coda cannot be complex (cannot be formed by consonant clusters). Let us analyse each transformation.\n\n*Vowel shortening\nLet us consider a CV:C stem. For the dubitative and passive (whose suffixes begin with a vowel), we would find words of the shape CV:CVC. Therefore, based on the basic syllabification rules, we would obtain CV:.CVC, the first syllable having a long vowel (but no coda), while the second has a coda (but only a short vowel).\n\nFor the other two forms, whose suffixes begin with a consonant, we would get words like *CV:CCVC, and, by syllabifying them, we would get *CV:C.CVC. Since the language does not allow CV:C syllables, the vowel shortening emerges, resulting in the\nform CVC.CVC.\n\nNotebulbonAn asterisk (*) before a word/sentence shows that it is not grammatically correct or is not attested in the\nlanguage.\n\n*Epenthesis\n\nFor CVCC type stems, in the case of dubitative and passive, whose suffixes begin with a vowel, we get words like CVCCVC, which, after syllabification, results in CVC.CVC. On the other hand, for the other two forms, we obtain *CVCCCVC, which, after syllabification, would result in *CVC.CCVC or *CVCC.CVC. In both cases, either the onset or the coda would have a consonant cluster, which is not allowed in the language. Therefore, an epenthetic vowel is added, resulting in the word CVCVeCCVC CV.CVeC.CVC.\n\nTherefore, we can posit that the two root change phenomena take place in order to avoid consonant clusters or super-heavy syllables (having both a long vowel and a coda).\n\nThese observations also help us understand why for blanks 6–7 and 12–13 the first vowel of the stem cannot be long (i.e., we cannot consider answers such to:yixal and to:yi:xal to be correct). If the first vowel of the stem were long, there would be no reason for it to be shortened in the imperative form since the environment in which it appears does not change.\n\nMoreover, if this is true, it means that the verbal stems (CV:C and CVCC) are not single words, since their structure would not be allowed.\n\nThe Evenki language is an extremely interesting example since it features three rare characteristics:\n\n- Vowel harmony is triggered by the first vowel and not the last vowel (as is common in Turkic and Mongolic languages).\n\n- The language has three different types of vowel harmony, each of them occurring in different contexts:\n\n- total vowel harmony (total assimilation) – used, for example, to\nform the plural;\n\n- harmony that only affects round vowels, based on backness;\n\n- harmony based on backness that affects all vowels, independent of roundness.\n\n- There is a neutral vowel (i) which neither triggers nor affects vowel harmony.","source_file":"chapters/04-phonology.tex","source_line_start":884,"license":"CC-BY-4.0","retrieval_text":"Phonology\nVowel harmony\nphonological rules and sound correspondences\nIt is interesting to consider the reason why the stem changes occur. These changes only happen for non-future and imperative, and these are the only forms for which the corresponding suffixes begin with a consonant. Thus, the epenthesis (rule 1b) is rather common, and it is used to avoid a three-consonant cluster, since this would impede the pronunciation.\n\nThe vowel shortening, on the other hand, highlights a much more interesting phenomenon which can be connected to syllable weight. Remember that in the previous chapter, we classified syllables into four types, depending on vowel length and on the existence of a coda. In this language, it would appear that syllables which have both a long vowel and a consonant are not allowed; moreover, the onset and the coda cannot be complex (cannot be formed by consonant clusters). Let us analyse each transformation.\n\n*Vowel shortening\nLet us consider a CV:C stem. For the dubitative and passive (whose suffixes begin with a vowel), we would find words of the shape CV:CVC. Therefore, based on the basic syllabification rules, we would obtain CV:.CVC, the first syllable having a long vowel (but no coda), while the second has a coda (but only a short vowel).\n\nFor the other two forms, whose suffixes begin with a consonant, we would get words like *CV:CCVC, and, by syllabifying them, we would get *CV:C.CVC. Since the language does not allow CV:C syllables, the vowel shortening emerges, resulting in the\nform CVC.CVC.\n\nNotebulbonAn asterisk (*) before a word/sentence shows that it is not grammatically correct or is not attested in the\nlanguage.\n\n*Epenthesis\n\nFor CVCC type stems, in the case of dubitative and passive, whose suffixes begin with a vowel, we get words like CVCCVC, which, after syllabification, results in CVC.CVC. On the other hand, for the other two forms, we obtain *CVCCCVC, which, after syllabification, would result in *CVC.CCVC or *CVCC.CVC. In both cases, either the onset or the coda would have a consonant cluster, which is not allowed in the language. Therefore, an epenthetic vowel is added, resulting in the word CVCVeCCVC CV.CVeC.CVC.\n\nTherefore, we can posit that the two root change phenomena take place in order to avoid consonant clusters or super-heavy syllables (having both a long vowel and a coda).\n\nThese observations also help us understand why for blanks 6–7 and 12–13 the first vowel of the stem cannot be long (i.e., we cannot consider answers such to:yixal and to:yi:xal to be correct). If the first vowel of the stem were long, there would be no reason for it to be shortened in the imperative form since the environment in which it appears does not change.\n\nMoreover, if this is true, it means that the verbal stems (CV:C and CVCC) are not single words, since their structure would not be allowed.\n\nThe Evenki language is an extremely interesting example since it features three rare characteristics:\n\n- Vowel harmony is triggered by the first vowel and not the last vowel (as is common in Turkic and Mongolic languages).\n\n- The language has three different types of vowel harmony, each of them occurring in different contexts:\n\n- total vowel harmony (total assimilation) – used, for example, to\nform the plural;\n\n- harmony that only affects round vowels, based on backness;\n\n- harmony based on backness that affects all vowels, independent of roundness.\n\n- There is a neutral vowel (i) which neither triggers nor affects vowel harmony."}
{"id":"book_method_c04_s06_p01","kind":"method","chapter":4,"chapter_title":"Phonology","section":6,"section_title":"Initial consonant mutationInitial consonant mutation","topic":"phonological rules and sound correspondences","text":"Initial consonant mutation is a phenomenon occurring, for example, in Celtic languages (Irish, Breton, Manx, etc.). In these languages, the first consonant of words can undergo certain transformations (called mutations) based on the grammatical context. The context in which the first consonant mutates is variable; it could be depending on possession (e.g., ‘my’ vs. ‘his’), on the numeral that follows (e.g., there might be one form for numerals 1-4 and another form for numerals 5-9), etc.\n\nIn linguistics problems, if the initial consonant mutation is featured, some mutation examples will be given (in specific contexts), and you will be asked to deduce the mutation other consonants undergo. Let us consider the following example from Welsh (it is known that c is pronounced like ‘c’ in ‘car’):\n\nContext 1 | Context 2 | Context 3\np | b | mh\nc | g | ngh\nt | |\n\nWe can see that in the first context, there are only voiceless stops, which, in the second context, become voiced (therefore, based on the pairs p – b and c – g, we can infer that the consonant t in the second context will mutate to d). In the third context, we notice that there are only nasal consonants (followed by h). Therefore, we can deduce that t will become nh in the third context.\n\nTherefore, in all contexts, the place of articulation is preserved, and it is only the manner of articulation and voicing that change (voiceless stop – voiced stop – voiceless nasal).\n\nAn example of a problem which features initial consonant mutation is 5.16.","source_file":"chapters/04-phonology.tex","source_line_start":1431,"license":"CC-BY-4.0","retrieval_text":"Phonology\nInitial consonant mutationInitial consonant mutation\nphonological rules and sound correspondences\nInitial consonant mutation is a phenomenon occurring, for example, in Celtic languages (Irish, Breton, Manx, etc.). In these languages, the first consonant of words can undergo certain transformations (called mutations) based on the grammatical context. The context in which the first consonant mutates is variable; it could be depending on possession (e.g., ‘my’ vs. ‘his’), on the numeral that follows (e.g., there might be one form for numerals 1-4 and another form for numerals 5-9), etc.\n\nIn linguistics problems, if the initial consonant mutation is featured, some mutation examples will be given (in specific contexts), and you will be asked to deduce the mutation other consonants undergo. Let us consider the following example from Welsh (it is known that c is pronounced like ‘c’ in ‘car’):\n\nContext 1 | Context 2 | Context 3\np | b | mh\nc | g | ngh\nt | |\n\nWe can see that in the first context, there are only voiceless stops, which, in the second context, become voiced (therefore, based on the pairs p – b and c – g, we can infer that the consonant t in the second context will mutate to d). In the third context, we notice that there are only nasal consonants (followed by h). Therefore, we can deduce that t will become nh in the third context.\n\nTherefore, in all contexts, the place of articulation is preserved, and it is only the manner of articulation and voicing that change (voiceless stop – voiced stop – voiceless nasal).\n\nAn example of a problem which features initial consonant mutation is 5.16."}
{"id":"book_method_c05_s01_p01","kind":"method","chapter":5,"chapter_title":"Noun and noun phrase","section":1,"section_title":"Introduction","topic":"noun morphology and noun phrases","text":"Before presenting and analysing the main types of problems, it is important to get accustomed to the following general concepts. Traditional notional definitions of grammatical categories include the following:\n\n- Noun: words that name specific objects, places, beings, etc.\n\n- Proper noun: refers to names of people or places (Giovanni, London, etc.)\n\n- Common noun: describes a class of entities (city, class, girl, cat, etc.).\n\nWe can also think about the ways in which words combine to form larger constituents and the roles these larger constituents play in syntactic structures. For instance:\n\n- Noun phrase: a phrase that has a noun as its head and performs various grammatical functions such as subject and direct object\n\nThus, we can define a sentence (S) as a combination of a noun phrase (NP) and a verb phrase (VP). We can write: S = NP + VP. For example:\n\nAndrew eats. (NP = Proper noun = Andrew, VP = V = eats)\nA boy eats those sandwiches. (NP = A boy, VP = V + NP = eats + those sandwiches).\nOther examples of noun phrases (the word in bold represents the head noun) can be: the good boy, the boy inside the house, the five green tables, etc.\n\nWe notice that the noun phrase includes the noun (the head component) and other words subordinated to it (definite article, possessives, adjectives, etc.). We usually refer to all of these subordinated components as modifiers.Sometimes the term determiner is also used in linguistics problems. In current theories, however, the term determiner refers strictly to the definite/indefinite article (the grammatical category of determination). Thus, the term modifier is usually preferred to refer to the elements of the noun phrase (adjectives, demonstratives, etc.).\n\nNotebulbon\nSome words can belong, depending on the sentence structure, to either the noun phrase or the verb phrase. For example, the sentences The boy inside the house eats and The boy eats inside the house differ only in terms of word order. In the former, inside the house is part of the noun phrase and modifies boy (the boy who is inside the house), while in the latter it refers to where the action takes place and modifies the verb, as part of the verb phrase.\n\n- Morpheme: the smallest linguistic unit with a meaning or a grammatical function. This is a broad term which includes different subtypes, such as stems (usually, the lexical stem of the word), prefixes, suffixes, etc. For example, the word undesirability is made up of four morphemes: the morpheme desire (which is also the stem), the morpheme -able (forming the adjective desirable), the morpheme un- (forming the adjective undesirable), and finally, the morpheme -ity (forming the noun undesirability).\n\n- Allomorph: a variant of a morpheme. In some situations, a morpheme can have two or more similar forms, which are chosen based on phonological considerations. For example, in the words impossible and intolerant, the prefixes im- and in-, although they look different, serve the same function, for which reason we can consider them to be the same morpheme. Moreover, from a phonological point of view, im- appears when the nasal assimilates to the place of articulation of the following sound (p in possible).\n\n- Affix: a morpheme added to the stem. It is also a rather broad term which includes different types of affixes, like suffix or prefix.\n\n- Prefix: affix placed at the beginning of the word.\n\n- Suffix: affix placed at the end of the word.\n\nAlthough English has only two types of affixes (prefixes and suffixes), other languages can have a much greater variety, as follows:\n\n- Infix: affix placed inside the stem.\n\nLet us consider the following examples from Bontoc:\n\nfikas | ‘strong’ | fumikas | ‘to be strong’\nkilad | ‘red’ | kumilad | ‘to be red’\npusi | ‘poor’ | pumusi | ‘to be poor’","source_file":"chapters/05-Noun.tex","source_line_start":5,"license":"CC-BY-4.0","retrieval_text":"Noun and noun phrase\nIntroduction\nnoun morphology and noun phrases\nBefore presenting and analysing the main types of problems, it is important to get accustomed to the following general concepts. Traditional notional definitions of grammatical categories include the following:\n\n- Noun: words that name specific objects, places, beings, etc.\n\n- Proper noun: refers to names of people or places (Giovanni, London, etc.)\n\n- Common noun: describes a class of entities (city, class, girl, cat, etc.).\n\nWe can also think about the ways in which words combine to form larger constituents and the roles these larger constituents play in syntactic structures. For instance:\n\n- Noun phrase: a phrase that has a noun as its head and performs various grammatical functions such as subject and direct object\n\nThus, we can define a sentence (S) as a combination of a noun phrase (NP) and a verb phrase (VP). We can write: S = NP + VP. For example:\n\nAndrew eats. (NP = Proper noun = Andrew, VP = V = eats)\nA boy eats those sandwiches. (NP = A boy, VP = V + NP = eats + those sandwiches).\nOther examples of noun phrases (the word in bold represents the head noun) can be: the good boy, the boy inside the house, the five green tables, etc.\n\nWe notice that the noun phrase includes the noun (the head component) and other words subordinated to it (definite article, possessives, adjectives, etc.). We usually refer to all of these subordinated components as modifiers.Sometimes the term determiner is also used in linguistics problems. In current theories, however, the term determiner refers strictly to the definite/indefinite article (the grammatical category of determination). Thus, the term modifier is usually preferred to refer to the elements of the noun phrase (adjectives, demonstratives, etc.).\n\nNotebulbon\nSome words can belong, depending on the sentence structure, to either the noun phrase or the verb phrase. For example, the sentences The boy inside the house eats and The boy eats inside the house differ only in terms of word order. In the former, inside the house is part of the noun phrase and modifies boy (the boy who is inside the house), while in the latter it refers to where the action takes place and modifies the verb, as part of the verb phrase.\n\n- Morpheme: the smallest linguistic unit with a meaning or a grammatical function. This is a broad term which includes different subtypes, such as stems (usually, the lexical stem of the word), prefixes, suffixes, etc. For example, the word undesirability is made up of four morphemes: the morpheme desire (which is also the stem), the morpheme -able (forming the adjective desirable), the morpheme un- (forming the adjective undesirable), and finally, the morpheme -ity (forming the noun undesirability).\n\n- Allomorph: a variant of a morpheme. In some situations, a morpheme can have two or more similar forms, which are chosen based on phonological considerations. For example, in the words impossible and intolerant, the prefixes im- and in-, although they look different, serve the same function, for which reason we can consider them to be the same morpheme. Moreover, from a phonological point of view, im- appears when the nasal assimilates to the place of articulation of the following sound (p in possible).\n\n- Affix: a morpheme added to the stem. It is also a rather broad term which includes different types of affixes, like suffix or prefix.\n\n- Prefix: affix placed at the beginning of the word.\n\n- Suffix: affix placed at the end of the word.\n\nAlthough English has only two types of affixes (prefixes and suffixes), other languages can have a much greater variety, as follows:\n\n- Infix: affix placed inside the stem.\n\nLet us consider the following examples from Bontoc:\n\nfikas | ‘strong’ | fumikas | ‘to be strong’\nkilad | ‘red’ | kumilad | ‘to be red’\npusi | ‘poor’ | pumusi | ‘to be poor’"}
{"id":"book_method_c05_s01_p02","kind":"method","chapter":5,"chapter_title":"Noun and noun phrase","section":1,"section_title":"Introduction","topic":"noun morphology and noun phrases","text":"fikas | ‘strong’ | fumikas | ‘to be strong’\nkilad | ‘red’ | kumilad | ‘to be red’\npusi | ‘poor’ | pumusi | ‘to be poor’\n\nWe can notice that the derivation of a verb from an adjective (e.g., ‘red’ ‘to be red’) is made by adding the letters um after the first letter of the adjective. Thus, -um- is an example of an infix.\n\n- Circumfix: is a type of affix with two components, one part that is added to the beginning of the word and another that is added to the end of the word.\n\nTo some extent, a circumfix can be considered as a combination of a prefix and a suffix. Nevertheless, there is a major difference between a circumfix and a prefix + suffix combination: the circumfix is indivisible (i.e., the part in the beginning has neither meaning nor function without the part at the end; both parts are needed in order to achieve a meaning or function), while a prefix + suffix combination represent two independent entities, each of them having their own function or meaning. An example of a circumfix can be found in Chickasaw, in which the negation is formed using the circumfix ik- -o added to the stem, with the additional property that the final vowel of the stem is dropped:\n\nchokma | ‘he is good’ | ikchokmo | ‘he is not good’\ntiwwi | ‘he opens’ | iktiwwo | ‘he does not open’\npalli | ‘it is hot’ | ikpallo | ‘it is not hot’\n\n- Transfix: is a type of discontinuous affix (resembling a combination of different infixes).\n\nTransfixes are generally associated with Semitic languages (Arabic, Hebrew, Maltese, etc.) in which, for example, the verb stems are discontinuous, consisting of two to four consonants. Thus, in Maltese, the stem of the verb ‘to write’ is k-t-b. In order to conjugate it, we need to add different vowels before, between or after the consonants of the stem. For example, kiteb = ‘he wrote’. So, in order to obtain the 3sg masculine past simple of the verb, we need to use the transfix -i-e-.\n\n- Sometimes, part of the stem is deleted. We will refer to the part of the word that is deleted as a disfix.\n\nLet us consider the following example in Alabama:\n\ntipsali ‘to break’ tipli ‘to break up’\n\nIn this case, the second word is formed from the first through the elision of the penultimate syllable of the stem (sa). Based on this example, we could in fact consider that the second word is the stem and sa is merely an infix added to form the verb to break. Let us consider another example from the same language: batatli – batli. The formation of the second word can be easily explained using our first theory, in which the penultimate syllable is dropped; nevertheless, if we try to use our second explanation, in which an infix is added, we notice that in the first example, the infix added is -sa-, while the second example uses the infix ta. Moreover, there seems to be no connection or pattern regarding the choice of the infix. As a result, the disfix is the reasonable explanation, since the deleted morpheme is not predictable (cannot be deduced), but rather it is intrinsic to the stem.\n\nMost of the linguistics problems focused on the noun phrase will include different ways of forming new words (either singular – plural, etc.) or will focus on how nouns interact with different modifiers.","source_file":"chapters/05-Noun.tex","source_line_start":5,"license":"CC-BY-4.0","retrieval_text":"Noun and noun phrase\nIntroduction\nnoun morphology and noun phrases\nfikas | ‘strong’ | fumikas | ‘to be strong’\nkilad | ‘red’ | kumilad | ‘to be red’\npusi | ‘poor’ | pumusi | ‘to be poor’\n\nWe can notice that the derivation of a verb from an adjective (e.g., ‘red’ ‘to be red’) is made by adding the letters um after the first letter of the adjective. Thus, -um- is an example of an infix.\n\n- Circumfix: is a type of affix with two components, one part that is added to the beginning of the word and another that is added to the end of the word.\n\nTo some extent, a circumfix can be considered as a combination of a prefix and a suffix. Nevertheless, there is a major difference between a circumfix and a prefix + suffix combination: the circumfix is indivisible (i.e., the part in the beginning has neither meaning nor function without the part at the end; both parts are needed in order to achieve a meaning or function), while a prefix + suffix combination represent two independent entities, each of them having their own function or meaning. An example of a circumfix can be found in Chickasaw, in which the negation is formed using the circumfix ik- -o added to the stem, with the additional property that the final vowel of the stem is dropped:\n\nchokma | ‘he is good’ | ikchokmo | ‘he is not good’\ntiwwi | ‘he opens’ | iktiwwo | ‘he does not open’\npalli | ‘it is hot’ | ikpallo | ‘it is not hot’\n\n- Transfix: is a type of discontinuous affix (resembling a combination of different infixes).\n\nTransfixes are generally associated with Semitic languages (Arabic, Hebrew, Maltese, etc.) in which, for example, the verb stems are discontinuous, consisting of two to four consonants. Thus, in Maltese, the stem of the verb ‘to write’ is k-t-b. In order to conjugate it, we need to add different vowels before, between or after the consonants of the stem. For example, kiteb = ‘he wrote’. So, in order to obtain the 3sg masculine past simple of the verb, we need to use the transfix -i-e-.\n\n- Sometimes, part of the stem is deleted. We will refer to the part of the word that is deleted as a disfix.\n\nLet us consider the following example in Alabama:\n\ntipsali ‘to break’ tipli ‘to break up’\n\nIn this case, the second word is formed from the first through the elision of the penultimate syllable of the stem (sa). Based on this example, we could in fact consider that the second word is the stem and sa is merely an infix added to form the verb to break. Let us consider another example from the same language: batatli – batli. The formation of the second word can be easily explained using our first theory, in which the penultimate syllable is dropped; nevertheless, if we try to use our second explanation, in which an infix is added, we notice that in the first example, the infix added is -sa-, while the second example uses the infix ta. Moreover, there seems to be no connection or pattern regarding the choice of the infix. As a result, the disfix is the reasonable explanation, since the deleted morpheme is not predictable (cannot be deduced), but rather it is intrinsic to the stem.\n\nMost of the linguistics problems focused on the noun phrase will include different ways of forming new words (either singular – plural, etc.) or will focus on how nouns interact with different modifiers."}
{"id":"book_method_c05_s02_p01","kind":"method","chapter":5,"chapter_title":"Noun and noun phrase","section":2,"section_title":"Basic principle of morphological analysis","topic":"noun morphology and noun phrases","text":"- If a single phonetic form has two distinct meanings (or functions), it must be analysed as representing two different morphemes.\n\nFor example, the morpheme -s in the words cats and sees could be considered, at first sight, to be a single morpheme, since it looks the same. Nevertheless, in each of the two words it serves different functions (in the first case it forms the plural of the noun, while in the second it forms the 3sg present tense). Therefore it needs to be treated as two different morphemes (the morpheme -s used to form the plural of the nouns, and the morpheme -s used for the 3sg present tense).\n\n- If the same function (or meaning) is associated with two or more phonetic forms, these different forms all represent the same morpheme and the choice of form in each case is usually predictable based on phonological or other considerations.\n\nFor example, the prefixes il-, im-, in-, ir- (illegal, impossible, intolerant, irresponsible) serve the same function, so we can consider them to be allomorphs of the same morpheme. Moreover, the choice of the allomorph can be predicted based on phonological considerations: if the stem starts with a liquid, the prefix will be iX-, where X is the first consonant of the stem; if the stem starts with a stop, the prefix will be iN-, where N is a nasal consonant that assimilates to the place of articulation from the first consonant in the stem.\n\nNotebulbonLinguistics problems featuring disfixes are extremely rare. Before considering the possibility of a disfix, make sure there is absolutely no correlation between that sound/morpheme and the shape or meaning of the stem.","source_file":"chapters/05-Noun.tex","source_line_start":93,"license":"CC-BY-4.0","retrieval_text":"Noun and noun phrase\nBasic principle of morphological analysis\nnoun morphology and noun phrases\n- If a single phonetic form has two distinct meanings (or functions), it must be analysed as representing two different morphemes.\n\nFor example, the morpheme -s in the words cats and sees could be considered, at first sight, to be a single morpheme, since it looks the same. Nevertheless, in each of the two words it serves different functions (in the first case it forms the plural of the noun, while in the second it forms the 3sg present tense). Therefore it needs to be treated as two different morphemes (the morpheme -s used to form the plural of the nouns, and the morpheme -s used for the 3sg present tense).\n\n- If the same function (or meaning) is associated with two or more phonetic forms, these different forms all represent the same morpheme and the choice of form in each case is usually predictable based on phonological or other considerations.\n\nFor example, the prefixes il-, im-, in-, ir- (illegal, impossible, intolerant, irresponsible) serve the same function, so we can consider them to be allomorphs of the same morpheme. Moreover, the choice of the allomorph can be predicted based on phonological considerations: if the stem starts with a liquid, the prefix will be iX-, where X is the first consonant of the stem; if the stem starts with a stop, the prefix will be iN-, where N is a nasal consonant that assimilates to the place of articulation from the first consonant in the stem.\n\nNotebulbonLinguistics problems featuring disfixes are extremely rare. Before considering the possibility of a disfix, make sure there is absolutely no correlation between that sound/morpheme and the shape or meaning of the stem."}
{"id":"book_method_c05_s03_p01","kind":"method","chapter":5,"chapter_title":"Noun and noun phrase","section":3,"section_title":"Variables of the noun","topic":"noun morphology and noun phrases","text":"When we talk about the variablesBy using the term noun variables, we refer to the grammatical categories specific to English (number, case, etc.), as well as to other distinctions which might be relevant in other languages (usually, of a semantic nature).\n\nAlthough all the grammatical categories of the noun are considered variables of the noun, not all variables are grammatical categories. For example, the distinction human or animate, to which we will refer in the following sections, is an important variable of the noun, but it cannot be considered a grammatical category. of the noun, we mean those features of the noun which can change in a linguistics problem and which it is best to identify when starting to solve the problem, in order to form an idea regarding what types of morphemes we are looking for. These variables can also be marked on other elements in the phrase/clause, e.g., on adjectives. This phenomenon is known as agreement.\n\nThe three common variables are: number (singular, dual, plural), gender (masculine, neuter, feminine etc.) which is a highly grammaticalised subtype of noun class. The way in which we usually realise that gender is relevant in a linguistics problem is by identifying different allomorphs. Thus, if we notice that a set of allomorphs appears solely with some stems and another set with other stems, we can assume in that language there are two classes of nouns (two genders), each of them having their own characteristic morphemes. Class and classifiers will be further discussed in the next section.\n\nThe number variable denotes how many items the noun refers to. The simplest distinction is singular (referring to one item) and plural (referring to more than one item). However, different languages might have other distinctions, among which the most common are dual (two items), trial (three items), and paucal (referring to a relatively small number of items. This would typically be translated into English as ‘a few’ and usually refers to less than ten items). Of course, there are also languages which do not mark nouns for number.\n\nAn interesting case regarding noun plurals is encountered in Dagaare.This phenomenon was featured in a problem by Ethan Chi (NACLO 2021). This language features an inherent plurality of the designated noun and, starting from it, it differentiates an unmarked form (a base form) and a marked form. Thus, if a noun represents an entity that is usually found alone (‘forehead’, ‘hat’), the unmarked form will be represented by the singular (while the plural will be the marked form). However, if the noun is usually found in pairs (‘leg’, ‘lung’, ‘shoe’) or groups (‘bee’), then the unmarked form is the plural (and the singular is denoted by the marked form).\n\nThere is one more variable of the noun that we have not mentioned yet: grammatical case. Case is an important feature of nouns and indicates the role the noun plays in the sentence. In some languages, nouns can be marked for many cases; for example, Uralic languages, which are known for their large number of cases, can have more than 20 cases – nominative, genitive, partitive, accusative, inessive, elative, illative, adessive, ablative, additive, egressive, comitative, terminative, abessive, translative, allative, essive, instructive, instrumental, dative, causal, sublative, superessive, delative, temporal, sociative. Most of these cases will be translated into English using different prepositions, for example, the instrumental case (which shows the object with which the action is performed) is, usually, translated using the preposition ‘with’ or ‘by’ in English (I write with a pen or I sew by hand). When solving a linguistics problem, assuming the instrumental is marked by the suffix -ok in the target language, we do not necessarily have to write “Instrumental = -ok\", but rather we can simply write; “‘with X’ = X-ok\", without using the name of the case.","source_file":"chapters/05-Noun.tex","source_line_start":275,"license":"CC-BY-4.0","retrieval_text":"Noun and noun phrase\nVariables of the noun\nnoun morphology and noun phrases\nWhen we talk about the variablesBy using the term noun variables, we refer to the grammatical categories specific to English (number, case, etc.), as well as to other distinctions which might be relevant in other languages (usually, of a semantic nature).\n\nAlthough all the grammatical categories of the noun are considered variables of the noun, not all variables are grammatical categories. For example, the distinction human or animate, to which we will refer in the following sections, is an important variable of the noun, but it cannot be considered a grammatical category. of the noun, we mean those features of the noun which can change in a linguistics problem and which it is best to identify when starting to solve the problem, in order to form an idea regarding what types of morphemes we are looking for. These variables can also be marked on other elements in the phrase/clause, e.g., on adjectives. This phenomenon is known as agreement.\n\nThe three common variables are: number (singular, dual, plural), gender (masculine, neuter, feminine etc.) which is a highly grammaticalised subtype of noun class. The way in which we usually realise that gender is relevant in a linguistics problem is by identifying different allomorphs. Thus, if we notice that a set of allomorphs appears solely with some stems and another set with other stems, we can assume in that language there are two classes of nouns (two genders), each of them having their own characteristic morphemes. Class and classifiers will be further discussed in the next section.\n\nThe number variable denotes how many items the noun refers to. The simplest distinction is singular (referring to one item) and plural (referring to more than one item). However, different languages might have other distinctions, among which the most common are dual (two items), trial (three items), and paucal (referring to a relatively small number of items. This would typically be translated into English as ‘a few’ and usually refers to less than ten items). Of course, there are also languages which do not mark nouns for number.\n\nAn interesting case regarding noun plurals is encountered in Dagaare.This phenomenon was featured in a problem by Ethan Chi (NACLO 2021). This language features an inherent plurality of the designated noun and, starting from it, it differentiates an unmarked form (a base form) and a marked form. Thus, if a noun represents an entity that is usually found alone (‘forehead’, ‘hat’), the unmarked form will be represented by the singular (while the plural will be the marked form). However, if the noun is usually found in pairs (‘leg’, ‘lung’, ‘shoe’) or groups (‘bee’), then the unmarked form is the plural (and the singular is denoted by the marked form).\n\nThere is one more variable of the noun that we have not mentioned yet: grammatical case. Case is an important feature of nouns and indicates the role the noun plays in the sentence. In some languages, nouns can be marked for many cases; for example, Uralic languages, which are known for their large number of cases, can have more than 20 cases – nominative, genitive, partitive, accusative, inessive, elative, illative, adessive, ablative, additive, egressive, comitative, terminative, abessive, translative, allative, essive, instructive, instrumental, dative, causal, sublative, superessive, delative, temporal, sociative. Most of these cases will be translated into English using different prepositions, for example, the instrumental case (which shows the object with which the action is performed) is, usually, translated using the preposition ‘with’ or ‘by’ in English (I write with a pen or I sew by hand). When solving a linguistics problem, assuming the instrumental is marked by the suffix -ok in the target language, we do not necessarily have to write “Instrumental = -ok\", but rather we can simply write; “‘with X’ = X-ok\", without using the name of the case."}
{"id":"book_method_c05_s03_p02","kind":"method","chapter":5,"chapter_title":"Noun and noun phrase","section":3,"section_title":"Variables of the noun","topic":"noun morphology and noun phrases","text":"There is one more variable of the noun that we have not mentioned yet: grammatical case. Case is an important feature of nouns and indicates the role the noun plays in the sentence. In some languages, nouns can be marked for many cases; for example, Uralic languages, which are known for their large number of cases, can have more than 20 cases – nominative, genitive, partitive, accusative, inessive, elative, illative, adessive, ablative, additive, egressive, comitative, terminative, abessive, translative, allative, essive, instructive, instrumental, dative, causal, sublative, superessive, delative, temporal, sociative. Most of these cases will be translated into English using different prepositions, for example, the instrumental case (which shows the object with which the action is performed) is, usually, translated using the preposition ‘with’ or ‘by’ in English (I write with a pen or I sew by hand). When solving a linguistics problem, assuming the instrumental is marked by the suffix -ok in the target language, we do not necessarily have to write “Instrumental = -ok\", but rather we can simply write; “‘with X’ = X-ok\", without using the name of the case.\n\nMoreover, most of the cases of the Uralic languages are locatives (they show the location). The elative case can be translated using ‘from +M’, illative = ‘in +M’, allative = ‘on +M’, adessive = ‘on -M’, etc, where M denotes movement. Thus, the cases +M are cases in which the object is moved in a direction (relative to the noun), while the -M cases are those in which the objects are already in that location.\n\nLet us consider, for example, the allative and adessive cases (both of them can be expressed in English using the preposition ‘on’, but they differ in terms of the parameter M). As such, in the sentence The apple is on the table, on links to the concept of adessive, since it shows a location without any movement involved. However, in the sentence He put the apple on the table, the apple is moved towards the table: in the beginning, the apple was not on the table, but at the end of the action it was; therefore, in this case, on links to the concept of allative. Notice that, because English has very little in the way of inflectional morphology, the equivalents of nouns that appear in the adessive or allative cases in other languages such as Finnish appear as the complements of prepositions in English. In other words, what Finnish expresses via the morphological structure of the noun, English expresses in its syntactic structure (i.e. as a prepositional phrase).\n\nBecause case is relevant in both morphology and syntax, we will discuss specific examples when they arise. For example, nominative, accusative, ergative, and absolutive, which we will discuss in Section morphoalign.","source_file":"chapters/05-Noun.tex","source_line_start":275,"license":"CC-BY-4.0","retrieval_text":"Noun and noun phrase\nVariables of the noun\nnoun morphology and noun phrases\nThere is one more variable of the noun that we have not mentioned yet: grammatical case. Case is an important feature of nouns and indicates the role the noun plays in the sentence. In some languages, nouns can be marked for many cases; for example, Uralic languages, which are known for their large number of cases, can have more than 20 cases – nominative, genitive, partitive, accusative, inessive, elative, illative, adessive, ablative, additive, egressive, comitative, terminative, abessive, translative, allative, essive, instructive, instrumental, dative, causal, sublative, superessive, delative, temporal, sociative. Most of these cases will be translated into English using different prepositions, for example, the instrumental case (which shows the object with which the action is performed) is, usually, translated using the preposition ‘with’ or ‘by’ in English (I write with a pen or I sew by hand). When solving a linguistics problem, assuming the instrumental is marked by the suffix -ok in the target language, we do not necessarily have to write “Instrumental = -ok\", but rather we can simply write; “‘with X’ = X-ok\", without using the name of the case.\n\nMoreover, most of the cases of the Uralic languages are locatives (they show the location). The elative case can be translated using ‘from +M’, illative = ‘in +M’, allative = ‘on +M’, adessive = ‘on -M’, etc, where M denotes movement. Thus, the cases +M are cases in which the object is moved in a direction (relative to the noun), while the -M cases are those in which the objects are already in that location.\n\nLet us consider, for example, the allative and adessive cases (both of them can be expressed in English using the preposition ‘on’, but they differ in terms of the parameter M). As such, in the sentence The apple is on the table, on links to the concept of adessive, since it shows a location without any movement involved. However, in the sentence He put the apple on the table, the apple is moved towards the table: in the beginning, the apple was not on the table, but at the end of the action it was; therefore, in this case, on links to the concept of allative. Notice that, because English has very little in the way of inflectional morphology, the equivalents of nouns that appear in the adessive or allative cases in other languages such as Finnish appear as the complements of prepositions in English. In other words, what Finnish expresses via the morphological structure of the noun, English expresses in its syntactic structure (i.e. as a prepositional phrase).\n\nBecause case is relevant in both morphology and syntax, we will discuss specific examples when they arise. For example, nominative, accusative, ergative, and absolutive, which we will discuss in Section morphoalign."}
{"id":"book_method_c05_s04_p01","kind":"method","chapter":5,"chapter_title":"Noun and noun phrase","section":4,"section_title":"Classifiers","topic":"noun morphology and noun phrases","text":"A classifier is a word (or an affix) which accompanies nouns and serves to classify them (thus talking about noun class). In most cases, the classification is made on semantic considerations. In Chinese, the classifier is mandatory between numeral and noun. Let us consider the following examples from Chinese:\n\n‘three dogs’ = 三只狗 ‘three cats’ = 三只猫 ‘five cats’ = 五只猫\n\nWe notice that the first character represents the number (三 = 3, 五 = 5), while the last one represents the noun (狗 = ‘dog’, 猫 = ‘cat’). The middle word represents a classifier and, in this case, it refers to small animals. Other semantic considerations for the choice of the classifier are, usually, related to the shape: long objects, flat objects, round objects, etc.","source_file":"chapters/05-Noun.tex","source_line_start":469,"license":"CC-BY-4.0","retrieval_text":"Noun and noun phrase\nClassifiers\nnoun morphology and noun phrases\nA classifier is a word (or an affix) which accompanies nouns and serves to classify them (thus talking about noun class). In most cases, the classification is made on semantic considerations. In Chinese, the classifier is mandatory between numeral and noun. Let us consider the following examples from Chinese:\n\n‘three dogs’ = 三只狗 ‘three cats’ = 三只猫 ‘five cats’ = 五只猫\n\nWe notice that the first character represents the number (三 = 3, 五 = 5), while the last one represents the noun (狗 = ‘dog’, 猫 = ‘cat’). The middle word represents a classifier and, in this case, it refers to small animals. Other semantic considerations for the choice of the classifier are, usually, related to the shape: long objects, flat objects, round objects, etc."}
{"id":"book_method_c05_s05_p01","kind":"method","chapter":5,"chapter_title":"Noun and noun phrase","section":5,"section_title":"Reduplication","topic":"noun morphology and noun phrases","text":"Reduplication represents the formation of new words by partially or totally doubling a morpheme or a part of a word. An example of total reduplication is plural formation in Indonesian, as we can see in the following examples: rumah (‘house’) – rumahrumah (‘houses’), ibu (‘mother’) – ibuibu (‘mothers’), lalat (‘fly’) – lalatlalat (‘flies’).\n\nWe can notice that in order to form the plural we simply repeat the whole word, so we are talking about a process of total reduplication.\n\nIn linguistics problems, total reduplication is seldom encountered, since it is easy to spot and analyse; therefore, in most cases, partial reduplication is preferred. The following examples show two such processes:\n\nMarshallese: | kagir (‘belt’) | kagirgir (‘to wear a belt’)\n| takin (‘socks’) | takinkin (‘to wear socks’)\n| |\nSamoan: | savali (‘he walks’) | savavali (‘they walk’)\n| alofa (‘he loves’) | alolofa (‘they love’)\n\nWe can easily notice that in the first example the last syllable is reduplicated, while in the second example, the second (or penultimate) syllable is reduplicated. Moreover, we notice that the reduplication process can have different roles, such as forming the plural of the noun (in Indonesian), forming the plural of the verb (transforming a verb from 3sg to 3pl in Samoan) or even transforming a noun into a verb (the pairs X – ‘to wear’ X, in Marshallese).\n\nNotebulbonAn alternative explanation for the aforementioned reduplication process can be “repeating the last three letters” (in Marshallese). Nevertheless, “the last three letters” do not represent a phonological or morphological unit of any sort, so this explanation is not very rigorous from a linguistic perspective.","source_file":"chapters/05-Noun.tex","source_line_start":586,"license":"CC-BY-4.0","retrieval_text":"Noun and noun phrase\nReduplication\nnoun morphology and noun phrases\nReduplication represents the formation of new words by partially or totally doubling a morpheme or a part of a word. An example of total reduplication is plural formation in Indonesian, as we can see in the following examples: rumah (‘house’) – rumahrumah (‘houses’), ibu (‘mother’) – ibuibu (‘mothers’), lalat (‘fly’) – lalatlalat (‘flies’).\n\nWe can notice that in order to form the plural we simply repeat the whole word, so we are talking about a process of total reduplication.\n\nIn linguistics problems, total reduplication is seldom encountered, since it is easy to spot and analyse; therefore, in most cases, partial reduplication is preferred. The following examples show two such processes:\n\nMarshallese: | kagir (‘belt’) | kagirgir (‘to wear a belt’)\n| takin (‘socks’) | takinkin (‘to wear socks’)\n| |\nSamoan: | savali (‘he walks’) | savavali (‘they walk’)\n| alofa (‘he loves’) | alolofa (‘they love’)\n\nWe can easily notice that in the first example the last syllable is reduplicated, while in the second example, the second (or penultimate) syllable is reduplicated. Moreover, we notice that the reduplication process can have different roles, such as forming the plural of the noun (in Indonesian), forming the plural of the verb (transforming a verb from 3sg to 3pl in Samoan) or even transforming a noun into a verb (the pairs X – ‘to wear’ X, in Marshallese).\n\nNotebulbonAn alternative explanation for the aforementioned reduplication process can be “repeating the last three letters” (in Marshallese). Nevertheless, “the last three letters” do not represent a phonological or morphological unit of any sort, so this explanation is not very rigorous from a linguistic perspective."}
{"id":"book_method_c05_s06_p01","kind":"method","chapter":5,"chapter_title":"Noun and noun phrase","section":6,"section_title":"Suppletion","topic":"noun morphology and noun phrases","text":"Another important phenomenon that might be featured in linguistics problems is suppletion. This generally refers to the situation in which one stem (so one morpheme with a semantic component) has two or more completely different forms (in most cases due to different etymologies). Suppletion can be noticed to a lesser extent in most languages, including English (good – better, bad – worse, go – went), French (être – sommes, aller – vais), Spanish (ir – voy), German (gut – besser), etc. We can notice that, in each of the cases, it is not simply a mutation of the stem (as for example in English: goose – geese or mouse – mice), but rather the two forms are completely distinct.\n\nThis phenomenon seldom appears in linguistics problems, which makes it extremely hard to notice and, similar to the disfixes, it must be used with caution since there might only be some phonological transformations.","source_file":"chapters/05-Noun.tex","source_line_start":608,"license":"CC-BY-4.0","retrieval_text":"Noun and noun phrase\nSuppletion\nnoun morphology and noun phrases\nAnother important phenomenon that might be featured in linguistics problems is suppletion. This generally refers to the situation in which one stem (so one morpheme with a semantic component) has two or more completely different forms (in most cases due to different etymologies). Suppletion can be noticed to a lesser extent in most languages, including English (good – better, bad – worse, go – went), French (être – sommes, aller – vais), Spanish (ir – voy), German (gut – besser), etc. We can notice that, in each of the cases, it is not simply a mutation of the stem (as for example in English: goose – geese or mouse – mice), but rather the two forms are completely distinct.\n\nThis phenomenon seldom appears in linguistics problems, which makes it extremely hard to notice and, similar to the disfixes, it must be used with caution since there might only be some phonological transformations."}
{"id":"book_method_c05_s07_p01","kind":"method","chapter":5,"chapter_title":"Noun and noun phrase","section":7,"section_title":"Expressing possession","topic":"noun morphology and noun phrases","text":"A common way of marking possession in the world's languages is to use the genitive case. This is usually expressed by a morpheme attached to the noun (and, sometimes, its dependents) that functions as the possessor.\n\nIn certain languages, often ones belonging to the Austronesian family, possession can be marked by an affix or an independent word. In these cases, the word used is a classifier (similar to those defined above). Moreover, in a multitude of languages, there is a dichotomy between alienable and inalienable possession. Inalienable possession refers to those objects one owns and which cannot be borrowed/lent or taken (in most cases, they include body parts and family members), while alienable possession refers to all the other cases. In most languages where this distinction is relevant, inalienable possession is formed by attaching the possessor morpheme directly to the noun, while alienable possession is formed using possession classes.","source_file":"chapters/05-Noun.tex","source_line_start":614,"license":"CC-BY-4.0","retrieval_text":"Noun and noun phrase\nExpressing possession\nnoun morphology and noun phrases\nA common way of marking possession in the world's languages is to use the genitive case. This is usually expressed by a morpheme attached to the noun (and, sometimes, its dependents) that functions as the possessor.\n\nIn certain languages, often ones belonging to the Austronesian family, possession can be marked by an affix or an independent word. In these cases, the word used is a classifier (similar to those defined above). Moreover, in a multitude of languages, there is a dichotomy between alienable and inalienable possession. Inalienable possession refers to those objects one owns and which cannot be borrowed/lent or taken (in most cases, they include body parts and family members), while alienable possession refers to all the other cases. In most languages where this distinction is relevant, inalienable possession is formed by attaching the possessor morpheme directly to the noun, while alienable possession is formed using possession classes."}
{"id":"book_method_c05_s08_p01","kind":"method","chapter":5,"chapter_title":"Noun and noun phrase","section":8,"section_title":"Colour terms","topic":"noun morphology and noun phrases","text":"A special focus is given to the problems that include noun phrases in which the adjectives are colour terms. For example, let us look at the following problem:\n\nStage I.\nW: Y, Br, R Bk: Bl, G\n\nStage II.\nW R: Y, Br Bk: Bl, G\n\nStage IIIa.\nW R: Br Y Bk: Bl, G\n*-+ 2\n\nStage IIIb.\nW R: Y, Br Bk G: Bl\n\nStage IV.\nW R: Br Y Bk G: Bl\n\nStage V.\nW R: Br Y Bk G Bl\n\nStage VI.\nW R Br Y Bk G Bl\n\nWe notice that the third stage has two variants: either ‘yellow’ splits away (from ‘red’ and ‘brown’), or ‘blue’ and ‘green’ (together) split from ‘black’; while in the fourth stage, both splits take place: ‘yellow’ splits (from ‘red’) and ‘blue’ and ‘green’ from ‘black’.\n\nThis fact can be useful in linguistics problems, considering the following aspect: historically speaking, it is likely that some languages had a limited number of colour terms and, in time, due to contact with other nations, they might have borrowed terms for other colours. For this reason, it is likely that, in some languages, some colour terms behave differently from others. In other words, it is likely that there are some basic colours (inherent to the language, which, according to the aforementioned hypothesis, should contain ‘white’, ‘black’, and ‘red’), which follow the usual declension of the language and other colour terms (borrowings from other languages) with a different flexion (usually diminished, or even inflexible).\n\nFor example, in Romanian, the basic colour terms have a full set of inflected forms, with four different forms (‘white’: alb – albă – albi – albe, ‘black’: negru – neagră – negri – negre), while other, new terms, are usually invariable (‘pink’: roz, ‘brown’: maro, ‘turquoise’: turcoaz, etc.).","source_file":"chapters/05-Noun.tex","source_line_start":1013,"license":"CC-BY-4.0","retrieval_text":"Noun and noun phrase\nColour terms\nnoun morphology and noun phrases\nA special focus is given to the problems that include noun phrases in which the adjectives are colour terms. For example, let us look at the following problem:\n\nStage I.\nW: Y, Br, R Bk: Bl, G\n\nStage II.\nW R: Y, Br Bk: Bl, G\n\nStage IIIa.\nW R: Br Y Bk: Bl, G\n*-+ 2\n\nStage IIIb.\nW R: Y, Br Bk G: Bl\n\nStage IV.\nW R: Br Y Bk G: Bl\n\nStage V.\nW R: Br Y Bk G Bl\n\nStage VI.\nW R Br Y Bk G Bl\n\nWe notice that the third stage has two variants: either ‘yellow’ splits away (from ‘red’ and ‘brown’), or ‘blue’ and ‘green’ (together) split from ‘black’; while in the fourth stage, both splits take place: ‘yellow’ splits (from ‘red’) and ‘blue’ and ‘green’ from ‘black’.\n\nThis fact can be useful in linguistics problems, considering the following aspect: historically speaking, it is likely that some languages had a limited number of colour terms and, in time, due to contact with other nations, they might have borrowed terms for other colours. For this reason, it is likely that, in some languages, some colour terms behave differently from others. In other words, it is likely that there are some basic colours (inherent to the language, which, according to the aforementioned hypothesis, should contain ‘white’, ‘black’, and ‘red’), which follow the usual declension of the language and other colour terms (borrowings from other languages) with a different flexion (usually diminished, or even inflexible).\n\nFor example, in Romanian, the basic colour terms have a full set of inflected forms, with four different forms (‘white’: alb – albă – albi – albe, ‘black’: negru – neagră – negri – negre), while other, new terms, are usually invariable (‘pink’: roz, ‘brown’: maro, ‘turquoise’: turcoaz, etc.)."}
{"id":"book_method_c06_s01_p01","kind":"method","chapter":6,"chapter_title":"Verb and verb phrase","section":1,"section_title":"Introduction","topic":"verb morphology and argument structure","text":"A number of general concepts are important for a discussion of verbs and verb phrases. Here are some notional definitions (i.e., those based on meaning) for key concepts:\n\n- verb: shows the action, existence, or state.In linguistics, there is a difference between verb and predicate, the former referring to the part of speech, while the latter to the part of a sentence. In linguistics problems, however, the term predicate is typically not used, and it is usually replaced by verb.\n\n- subject: typically shows who or what performs the action.\n\n- (direct) object: typically shows the person or object which is acted upon by the subject.\n\n- indirect object: shows the entity upon which the action is reflected indirectly.\n\nWe can also think about connections between verbs and noun phrases. For example:\n\n- transitive and intransitive verb: a transitive verb is one which has a direct object (I broke the glass) while an intransitive verb is one which has no direct object (The baby yawned). Note that in English we cannot say *I broke or *The baby yawned his mouth. Some verbs can vary in whether they take a direct object: thus, the verb to eat can be both transitive (She is eating a pizza – transitive use of eat, with a pizza as the direct object) and intransitive (She is eating).","source_file":"chapters/06-Verb.tex","source_line_start":5,"license":"CC-BY-4.0","retrieval_text":"Verb and verb phrase\nIntroduction\nverb morphology and argument structure\nA number of general concepts are important for a discussion of verbs and verb phrases. Here are some notional definitions (i.e., those based on meaning) for key concepts:\n\n- verb: shows the action, existence, or state.In linguistics, there is a difference between verb and predicate, the former referring to the part of speech, while the latter to the part of a sentence. In linguistics problems, however, the term predicate is typically not used, and it is usually replaced by verb.\n\n- subject: typically shows who or what performs the action.\n\n- (direct) object: typically shows the person or object which is acted upon by the subject.\n\n- indirect object: shows the entity upon which the action is reflected indirectly.\n\nWe can also think about connections between verbs and noun phrases. For example:\n\n- transitive and intransitive verb: a transitive verb is one which has a direct object (I broke the glass) while an intransitive verb is one which has no direct object (The baby yawned). Note that in English we cannot say *I broke or *The baby yawned his mouth. Some verbs can vary in whether they take a direct object: thus, the verb to eat can be both transitive (She is eating a pizza – transitive use of eat, with a pizza as the direct object) and intransitive (She is eating)."}
{"id":"book_method_c06_s02_p01","kind":"method","chapter":6,"chapter_title":"Verb and verb phrase","section":2,"section_title":"Variables of the verb","topic":"verb morphology and argument structure","text":"By variables, we mean those parameters which change in the problem. For example, comparing the sentences:\n\nHe eats. I ate.\n\nwe notice that the variables are subject (he vs. I), tense (present vs. past), but not the verb itself (both sentences have the same verb - eat).\n\nGenerally, these variables can be classified into three categories:\n\n- TAM (Tense, Aspect, Mood)\n\n- Arguments (S, S+O, S+O+O)\n\n- Others","source_file":"chapters/06-Verb.tex","source_line_start":22,"license":"CC-BY-4.0","retrieval_text":"Verb and verb phrase\nVariables of the verb\nverb morphology and argument structure\nBy variables, we mean those parameters which change in the problem. For example, comparing the sentences:\n\nHe eats. I ate.\n\nwe notice that the variables are subject (he vs. I), tense (present vs. past), but not the verb itself (both sentences have the same verb - eat).\n\nGenerally, these variables can be classified into three categories:\n\n- TAM (Tense, Aspect, Mood)\n\n- Arguments (S, S+O, S+O+O)\n\n- Others"}
{"id":"book_method_c06_s03_p01","kind":"method","chapter":6,"chapter_title":"Verb and verb phrase","section":3,"section_title":"TAM (Tense, Aspect, Mood)","topic":"verb morphology and argument structure","text":"Tense\nTense is a grammatical category, related to the concept of time. Languages often have three subtypes of tense: past (often used when the action or state denoted by the verb occurs before the moment of speech), present (action happens while speaking) and future (action will take place after the moment of speech).\n\nAlthough most of the Indo-European languages have all of these tenses, there are languages which only have two distinct tenses. Thus, there are languages which only distinguish past and non-past, such as Arabic and Japanese (non-past refers to everything that is not past, thus representing present and future) or languages which distinguish future and non-future, such as Greenlandic or Nivkh (where non-future represents past and present), or even languages that distinguish only present and non-present (one example is the constructed language Ithkuil). Moreover, there can be languages which make no tense distinctions at all, such as Chinese or Dyirbal.\n\nAlthough there are only three broad categories of tense, some languages can have many more actual tenses, mainly depending on the specific moment at which the action occurred. For example, the Yagua language has five different past tense markers:This phenomenon was featured in a problem by Vlad A. Neacșu (RoLO 2018).\n\n- -jásiy – for actions which took place a couple of hours ago (in the same day);\n\n- -jay – for actions occurring one day ago;\n\n- siy – one week to one month ago;\n\n- tíy – one or two months to one or two years ago;\n\n- -jada – more than two years ago (it is also called distant past or legendary past).\n\nAdditionally, some languages can have specific tenses, e.g., specific for actions occurring one day ago, the previous day (hesternal tense) or the following day (crastinal tense). There can even be pre-hesternal and post-crastinal tenses, referring to actions occurring two days before/after the moment of speech.\n\nSome languages also have hodiernal tenses, which are specific to actions occurring on the same day as the moment of speech (today). These can be past tenses (actions happening earlier today) or future tenses (actions happening later today).\n\nTherefore, returning to the Yagua example, we can define the marker a. (-jásiy) as a past hodiernal tense, while b. (-jay) can be considered a hesternal tense marker.\n\nAspect\n\nAspect shows the evolution in time of an action, state or event, independent of the moment of speech. The most common aspects are the perfective and imperfective.\n\nPerfective aspect is used when the event denoted by the verb is bounded. The imperfective is used when the event is seen as unfolding, or when the event is repeated/habitual. Depending on the language, there can also be other aspects such as:\n\n- Progressive = action is unfolding (progressing).\n\n- Semelfactive = action is short-term.\n\n- Accidental = action is done by mistake.","source_file":"chapters/06-Verb.tex","source_line_start":38,"license":"CC-BY-4.0","retrieval_text":"Verb and verb phrase\nTAM (Tense, Aspect, Mood)\nverb morphology and argument structure\nTense\nTense is a grammatical category, related to the concept of time. Languages often have three subtypes of tense: past (often used when the action or state denoted by the verb occurs before the moment of speech), present (action happens while speaking) and future (action will take place after the moment of speech).\n\nAlthough most of the Indo-European languages have all of these tenses, there are languages which only have two distinct tenses. Thus, there are languages which only distinguish past and non-past, such as Arabic and Japanese (non-past refers to everything that is not past, thus representing present and future) or languages which distinguish future and non-future, such as Greenlandic or Nivkh (where non-future represents past and present), or even languages that distinguish only present and non-present (one example is the constructed language Ithkuil). Moreover, there can be languages which make no tense distinctions at all, such as Chinese or Dyirbal.\n\nAlthough there are only three broad categories of tense, some languages can have many more actual tenses, mainly depending on the specific moment at which the action occurred. For example, the Yagua language has five different past tense markers:This phenomenon was featured in a problem by Vlad A. Neacșu (RoLO 2018).\n\n- -jásiy – for actions which took place a couple of hours ago (in the same day);\n\n- -jay – for actions occurring one day ago;\n\n- siy – one week to one month ago;\n\n- tíy – one or two months to one or two years ago;\n\n- -jada – more than two years ago (it is also called distant past or legendary past).\n\nAdditionally, some languages can have specific tenses, e.g., specific for actions occurring one day ago, the previous day (hesternal tense) or the following day (crastinal tense). There can even be pre-hesternal and post-crastinal tenses, referring to actions occurring two days before/after the moment of speech.\n\nSome languages also have hodiernal tenses, which are specific to actions occurring on the same day as the moment of speech (today). These can be past tenses (actions happening earlier today) or future tenses (actions happening later today).\n\nTherefore, returning to the Yagua example, we can define the marker a. (-jásiy) as a past hodiernal tense, while b. (-jay) can be considered a hesternal tense marker.\n\nAspect\n\nAspect shows the evolution in time of an action, state or event, independent of the moment of speech. The most common aspects are the perfective and imperfective.\n\nPerfective aspect is used when the event denoted by the verb is bounded. The imperfective is used when the event is seen as unfolding, or when the event is repeated/habitual. Depending on the language, there can also be other aspects such as:\n\n- Progressive = action is unfolding (progressing).\n\n- Semelfactive = action is short-term.\n\n- Accidental = action is done by mistake."}
{"id":"book_method_c06_s03_p02","kind":"method","chapter":6,"chapter_title":"Verb and verb phrase","section":3,"section_title":"TAM (Tense, Aspect, Mood)","topic":"verb morphology and argument structure","text":"- Accidental = action is done by mistake.\n\nThere are a lot of different aspects and their meaning can usually be inferred from their name (punctual, resumptive, intensive, attenuative, moderative, experiential, durative, pausative, terminative, episodic, generic, habitual, discontinuous, prospective, intentional, etc.). Memorising the names of all these aspects is not necessary, but it is important that you get used to the different meanings these aspects convey. Moreover, all of these aspects (not all of which exist in English so they may be expressed in more roundabout ways) need to be translated into English in a linguistics problem (most likely through a specific phrasing or by using an adverb). Therefore, instead of referring to the aspect itself, when solving a linguistics problem, you can simply write “The structures which are translated as...”. For example, the durative aspect can be translated into English by phrases such as ‘for a while’. Therefore, even if you do not know the name of the aspect (durative), it is enough to realise that there is a distinction between, for example, You ate and You ate for a while, in which case you can explain the sense of the durative aspect marker through the phrase for a while. Other examples are: potential aspect probably, prospective aspect I'm getting ready to..., etc.\n\nRememberbulbIn English, at least at school, aspect is not often talked about and the idea of “tense” refers, in fact, to a combination of tense and aspect. English has two main aspects: perfect aspect (e.g., I have/had eaten a sandwich, as defined above) and progressive aspect (e.g., I am/was eating a sandwich, sometimes known as continuous aspect).\n\nMood\n\nMood is an inflectional category related to the concept of modality, which is often connected with the speaker's attitude towards the message being transmitted (whether it is a fact, a wish, a command, etc.). Moods are classified into two broad classes: realis moods which show that something is a statement or a fact; and irrealis moods which refer to actions which have not happened (or will certainly not happen).\n\nRealis moods\n\n- Indicative mood: It is the most common mood and it is used for statements of fact. It is considered that all situations in a particular language that cannot be categorised as another mood will be classified as belonging to the indicative mood.\n\n- Certain languages can have another realis mood, a special mood which is used solely for general truths, as in the examples Fish swim or Chickens have two legs.\n\nIrrealis mood\n\nThere are many different subtypes of irrealis. As with the Uralic cases listed in Section sec:variables-of-the-noun, there is no need for you to remember all these subtypes. However, some subtypes appear more frequently in linguistics problems and we briefly describe these below.\n\n- Subjunctive mood: marks imaginary/hypothetical events, as well as opinions and emotions. It is the main irrealis mood and it represents an “umbrella term” for all the instances in which one language does not have another mood to express that attitude.\n\n- Conditional mood: the action is conditioned (by another action).\n\n- Optative mood: shows desires or wishes.\n\n- Imperative mood: direct commands/request/interdictions.\n\n- Jussive mood: similar to the imperative, but it expresses commands towards a third person, not present. Since in English this mood is not expressed directly by the verb, we usually use the subjunctive mood to express a similar meaning, e.g., I asked that he cook.\n\nVerbal expression of modality","source_file":"chapters/06-Verb.tex","source_line_start":38,"license":"CC-BY-4.0","retrieval_text":"Verb and verb phrase\nTAM (Tense, Aspect, Mood)\nverb morphology and argument structure\n- Accidental = action is done by mistake.\n\nThere are a lot of different aspects and their meaning can usually be inferred from their name (punctual, resumptive, intensive, attenuative, moderative, experiential, durative, pausative, terminative, episodic, generic, habitual, discontinuous, prospective, intentional, etc.). Memorising the names of all these aspects is not necessary, but it is important that you get used to the different meanings these aspects convey. Moreover, all of these aspects (not all of which exist in English so they may be expressed in more roundabout ways) need to be translated into English in a linguistics problem (most likely through a specific phrasing or by using an adverb). Therefore, instead of referring to the aspect itself, when solving a linguistics problem, you can simply write “The structures which are translated as...”. For example, the durative aspect can be translated into English by phrases such as ‘for a while’. Therefore, even if you do not know the name of the aspect (durative), it is enough to realise that there is a distinction between, for example, You ate and You ate for a while, in which case you can explain the sense of the durative aspect marker through the phrase for a while. Other examples are: potential aspect probably, prospective aspect I'm getting ready to..., etc.\n\nRememberbulbIn English, at least at school, aspect is not often talked about and the idea of “tense” refers, in fact, to a combination of tense and aspect. English has two main aspects: perfect aspect (e.g., I have/had eaten a sandwich, as defined above) and progressive aspect (e.g., I am/was eating a sandwich, sometimes known as continuous aspect).\n\nMood\n\nMood is an inflectional category related to the concept of modality, which is often connected with the speaker's attitude towards the message being transmitted (whether it is a fact, a wish, a command, etc.). Moods are classified into two broad classes: realis moods which show that something is a statement or a fact; and irrealis moods which refer to actions which have not happened (or will certainly not happen).\n\nRealis moods\n\n- Indicative mood: It is the most common mood and it is used for statements of fact. It is considered that all situations in a particular language that cannot be categorised as another mood will be classified as belonging to the indicative mood.\n\n- Certain languages can have another realis mood, a special mood which is used solely for general truths, as in the examples Fish swim or Chickens have two legs.\n\nIrrealis mood\n\nThere are many different subtypes of irrealis. As with the Uralic cases listed in Section sec:variables-of-the-noun, there is no need for you to remember all these subtypes. However, some subtypes appear more frequently in linguistics problems and we briefly describe these below.\n\n- Subjunctive mood: marks imaginary/hypothetical events, as well as opinions and emotions. It is the main irrealis mood and it represents an “umbrella term” for all the instances in which one language does not have another mood to express that attitude.\n\n- Conditional mood: the action is conditioned (by another action).\n\n- Optative mood: shows desires or wishes.\n\n- Imperative mood: direct commands/request/interdictions.\n\n- Jussive mood: similar to the imperative, but it expresses commands towards a third person, not present. Since in English this mood is not expressed directly by the verb, we usually use the subjunctive mood to express a similar meaning, e.g., I asked that he cook.\n\nVerbal expression of modality"}
{"id":"book_method_c06_s03_p03","kind":"method","chapter":6,"chapter_title":"Verb and verb phrase","section":3,"section_title":"TAM (Tense, Aspect, Mood)","topic":"verb morphology and argument structure","text":"Verbal expression of modality\n\nIn certain languages, modal verbs are used to express different kinds of modality. In many languages, modal verbs are considered auxiliary verbs. This means that they are accompanied by a ``lexical verb'', i.e., a verb with semantic content. In English, the central modal verbs (such as must and should) take the plain form of the verb as their complement (e.g., I must leave, you should eat). They can also combine with aspect markers (he must be working, she should have stayed): notice here that it is the following verb (the aspect marker) that is unchanged; the form of the lexical verb is determined by the aspect marker (be + Ving, have + Ved).\n\nOther verbs can also be used to express modality, and such verbs often take the to-infinitive as their complement. For example, in the sentence I want to go, want is the verb which shows the modality (the desire), while go is the verb with semantic content, showing the action I want/desire to perform. Other such patterns in English include wish to V, try to V, etc.\n\nThe verbs in bold in sentences like I want to swim and I tried to swim are sometimes known as catenative verbs because they can form a sequence or chain of verbs (catena is Latin for ‘chain’), as in I want to try to swim. Catenative verbs take a non-finite verb (e.g., an infinitive or participle) as their complement in English. Such catenative verbs can express modality (I need to leave) and aspect (He kept swimming), and since catenative verbs can combine, a sentence can involve the marking of both modality and aspect (I need to keep swimming).\n\nThis concept is relevant because it involves an interdependency between the two verbs, which are grammatically and semantically interconnected. Thus, in linguistics problems, if sequences of verbs occur, we need to pay attention to the following potential parameters: the order of the modal/catenative verb and the semantic verb (whether it comes before or after it, or whether there are other words or morphemes in between), as well as which of the two verbs gets conjugated (in some languages, only the auxiliary verbs are conjugated; in others, only the lexical verb; and in yet others, both the auxiliary and the lexical verb are conjugated).\n\nEvidentiality\n\nEvidentiality, unlike tense, aspect, and mood, shows the way in which the uttered information was discovered or, in other words, what evidence there is for the transmitted information. For example, in Pomo, there are four types of evidentiality, each of them having its own marker:\n\n- Visual: the speaker witnessed the action;\n\n- Sensorial non-visual: the speaker felt (by hearing, smelling, etc.) something that pointed towards the action. For example, the sentence The kids fought can receive a sensorial non-visual evidentiality marker to point out that the speaker has not seen the children fighting, but heard them;\n\n- Inferential: the speaker did not witness the action but was able to see its consequence or result. For example, the sentence The man cooked the fish can carry an inferential evidentiality marker to show that the speaker has not seen the fish being cooked (the process of cooking), but, for example, saw someone holding a plate with the cooked fish (thus, being able to infer that, at some point, the fish went through the cooking process).\n\n- Reportative: the speaker found out the information from someone else.\n\nFor example, the English sentence It rained could be translated into Pomo in four different ways, depending on the source of the evidence. Visual – the speaker sees that it's raining; sensorial non-visual – they hear the rain; inferential – they notice it is wet outside; reportative – someone tells them it rained.\n\nOther evidential contrasts can be:","source_file":"chapters/06-Verb.tex","source_line_start":38,"license":"CC-BY-4.0","retrieval_text":"Verb and verb phrase\nTAM (Tense, Aspect, Mood)\nverb morphology and argument structure\nVerbal expression of modality\n\nIn certain languages, modal verbs are used to express different kinds of modality. In many languages, modal verbs are considered auxiliary verbs. This means that they are accompanied by a ``lexical verb'', i.e., a verb with semantic content. In English, the central modal verbs (such as must and should) take the plain form of the verb as their complement (e.g., I must leave, you should eat). They can also combine with aspect markers (he must be working, she should have stayed): notice here that it is the following verb (the aspect marker) that is unchanged; the form of the lexical verb is determined by the aspect marker (be + Ving, have + Ved).\n\nOther verbs can also be used to express modality, and such verbs often take the to-infinitive as their complement. For example, in the sentence I want to go, want is the verb which shows the modality (the desire), while go is the verb with semantic content, showing the action I want/desire to perform. Other such patterns in English include wish to V, try to V, etc.\n\nThe verbs in bold in sentences like I want to swim and I tried to swim are sometimes known as catenative verbs because they can form a sequence or chain of verbs (catena is Latin for ‘chain’), as in I want to try to swim. Catenative verbs take a non-finite verb (e.g., an infinitive or participle) as their complement in English. Such catenative verbs can express modality (I need to leave) and aspect (He kept swimming), and since catenative verbs can combine, a sentence can involve the marking of both modality and aspect (I need to keep swimming).\n\nThis concept is relevant because it involves an interdependency between the two verbs, which are grammatically and semantically interconnected. Thus, in linguistics problems, if sequences of verbs occur, we need to pay attention to the following potential parameters: the order of the modal/catenative verb and the semantic verb (whether it comes before or after it, or whether there are other words or morphemes in between), as well as which of the two verbs gets conjugated (in some languages, only the auxiliary verbs are conjugated; in others, only the lexical verb; and in yet others, both the auxiliary and the lexical verb are conjugated).\n\nEvidentiality\n\nEvidentiality, unlike tense, aspect, and mood, shows the way in which the uttered information was discovered or, in other words, what evidence there is for the transmitted information. For example, in Pomo, there are four types of evidentiality, each of them having its own marker:\n\n- Visual: the speaker witnessed the action;\n\n- Sensorial non-visual: the speaker felt (by hearing, smelling, etc.) something that pointed towards the action. For example, the sentence The kids fought can receive a sensorial non-visual evidentiality marker to point out that the speaker has not seen the children fighting, but heard them;\n\n- Inferential: the speaker did not witness the action but was able to see its consequence or result. For example, the sentence The man cooked the fish can carry an inferential evidentiality marker to show that the speaker has not seen the fish being cooked (the process of cooking), but, for example, saw someone holding a plate with the cooked fish (thus, being able to infer that, at some point, the fish went through the cooking process).\n\n- Reportative: the speaker found out the information from someone else.\n\nFor example, the English sentence It rained could be translated into Pomo in four different ways, depending on the source of the evidence. Visual – the speaker sees that it's raining; sensorial non-visual – they hear the rain; inferential – they notice it is wet outside; reportative – someone tells them it rained.\n\nOther evidential contrasts can be:"}
{"id":"book_method_c06_s03_p04","kind":"method","chapter":6,"chapter_title":"Verb and verb phrase","section":3,"section_title":"TAM (Tense, Aspect, Mood)","topic":"verb morphology and argument structure","text":"Other evidential contrasts can be:\n\n- Witness vs. non-witness: the speaker witnessed the action (the information was retrieved through direct observation) or not. This type of evidentiality occurs in Turkish, where there are two types of past, called seen past (görülen geçmiş zaman, witness evidentiality marker) and heard past (duyulan geçmiş zaman, non-witness evidential).\n\n- First-hand vs. second-hand vs. third-hand: first-hand information is equivalent to directly observed information (witness); second-hand information is used to show that the speaker found out the information from someone else (who witnessed the action), while third-hand information indicates that the speaker found out the information from another person (who, in turn, found it out from a third person who witnessed the action).\n\nAs in the case of TAM categories, there can be other evidential markers and, depending on the language, each of these categories can be further divided into subcategories. For example, the inferential evidential can have the following subtypes: information deduced based on direct evidence (seeing the result of the action), information deduced based on general knowledge, information deduced (or inferred) based on the speaker's past experiences in similar situations, etc.\n\nMoreover, evidentiality can also be combined with tense, aspect, and mood, for which reason some linguists prefer using the abbreviation TAME (instead of TAM): tense, aspect, mood, evidentiality.","source_file":"chapters/06-Verb.tex","source_line_start":38,"license":"CC-BY-4.0","retrieval_text":"Verb and verb phrase\nTAM (Tense, Aspect, Mood)\nverb morphology and argument structure\nOther evidential contrasts can be:\n\n- Witness vs. non-witness: the speaker witnessed the action (the information was retrieved through direct observation) or not. This type of evidentiality occurs in Turkish, where there are two types of past, called seen past (görülen geçmiş zaman, witness evidentiality marker) and heard past (duyulan geçmiş zaman, non-witness evidential).\n\n- First-hand vs. second-hand vs. third-hand: first-hand information is equivalent to directly observed information (witness); second-hand information is used to show that the speaker found out the information from someone else (who witnessed the action), while third-hand information indicates that the speaker found out the information from another person (who, in turn, found it out from a third person who witnessed the action).\n\nAs in the case of TAM categories, there can be other evidential markers and, depending on the language, each of these categories can be further divided into subcategories. For example, the inferential evidential can have the following subtypes: information deduced based on direct evidence (seeing the result of the action), information deduced based on general knowledge, information deduced (or inferred) based on the speaker's past experiences in similar situations, etc.\n\nMoreover, evidentiality can also be combined with tense, aspect, and mood, for which reason some linguists prefer using the abbreviation TAME (instead of TAM): tense, aspect, mood, evidentiality."}
{"id":"book_method_c06_s04_p01","kind":"method","chapter":6,"chapter_title":"Verb and verb phrase","section":4,"section_title":"Arguments","topic":"verb morphology and argument structure","text":"In many linguistics problems, we need to identify the arguments of the verb. These are often the subject (S) and object(s) (O) of the verb. In some cases, we also distinguish between the subject of an intransitive verb and that of a transitive verb. This will be discussed more thoroughly in Section morphoalign.\n\nIn verb (phrase) problems, S and O are often expressed as pronouns in the English translations, though they may be directly attached to the verb as affixes in the target language. Depending on the difficulty of the problem, we can have different situations:\n\n- Easy problems: subject and object are marked through distinct, independent affixes.\n\n- Medium problems: in which S and O are either combined into a single affix, or they undergo certain phonological changes.\n\n- Hard problems: in which S and O are not expressed uniquely, but rather through a combination of affixes.\n\nFor more information about case alignment, see Section morphoalign.","source_file":"chapters/06-Verb.tex","source_line_start":132,"license":"CC-BY-4.0","retrieval_text":"Verb and verb phrase\nArguments\nverb morphology and argument structure\nIn many linguistics problems, we need to identify the arguments of the verb. These are often the subject (S) and object(s) (O) of the verb. In some cases, we also distinguish between the subject of an intransitive verb and that of a transitive verb. This will be discussed more thoroughly in Section morphoalign.\n\nIn verb (phrase) problems, S and O are often expressed as pronouns in the English translations, though they may be directly attached to the verb as affixes in the target language. Depending on the difficulty of the problem, we can have different situations:\n\n- Easy problems: subject and object are marked through distinct, independent affixes.\n\n- Medium problems: in which S and O are either combined into a single affix, or they undergo certain phonological changes.\n\n- Hard problems: in which S and O are not expressed uniquely, but rather through a combination of affixes.\n\nFor more information about case alignment, see Section morphoalign."}
{"id":"book_method_c06_s06_p01","kind":"method","chapter":6,"chapter_title":"Verb and verb phrase","section":6,"section_title":"Patterns for arguments","topic":"verb morphology and argument structure","text":"If the arguments (subject and object) are not marked by a single morpheme, it is possible that they will be marked by a combination of morphemes. Thus, they can be split into individual morphemes representing the person (1, 2, or 3), number (singular, dual, plural), or gender (masculine, feminine). Therefore, if an argument cannot be encountered as a single morpheme, we need to check whether there are any correlations between individual variables. Moreover, it is rather common for the third person to be unmarked, i.e., not to have a specific morpheme.","source_file":"chapters/06-Verb.tex","source_line_start":570,"license":"CC-BY-4.0","retrieval_text":"Verb and verb phrase\nPatterns for arguments\nverb morphology and argument structure\nIf the arguments (subject and object) are not marked by a single morpheme, it is possible that they will be marked by a combination of morphemes. Thus, they can be split into individual morphemes representing the person (1, 2, or 3), number (singular, dual, plural), or gender (masculine, feminine). Therefore, if an argument cannot be encountered as a single morpheme, we need to check whether there are any correlations between individual variables. Moreover, it is rather common for the third person to be unmarked, i.e., not to have a specific morpheme."}
{"id":"book_method_c06_s07_p01","kind":"method","chapter":6,"chapter_title":"Verb and verb phrase","section":7,"section_title":"Pronoun hierarchy","topic":"verb morphology and argument structure","text":"In many languages that we are familiar with, the subject and object are distinguished solely based on the word or morpheme order. For example, the difference between The dog sees the cat and The cat sees the dog is strictly due to the word order. Since the subject is, generally, the first in the sentence, we know that in the first sentence the dog is the subject and the cat is the object. Swapping the positions of the two, we also change their roles: so that in the second sentence, the cat becomes the subject, while the dog is the object.\n\nEven in the problems we have solved up to now, the same rules have applied. If we look back at Problem 6.1 (Swahili), we remember that the subject was always placed first, while the object was placed third. Thus, the structure ninawuona (ni-na-wu-ona = 1sg-present-2sg-‘see’) was translated as ‘I see yousg’, but, if we reverse the order of morphemes I and III (wunaniona), we get ‘yousg see me’.\n\nNevertheless, some languages work quite differently. The order of the two pronoun morphemes can be completely independent of their roles (subject or object), but rather depends on a predefined hierarchy of the persons together with another morpheme which shows whether the subject and object follow that hierarchy or not. The most common pronominal hierarchies are 1 > 2 > 3 (meaning that the 1st person is prioritised over the 2nd, which is then prioritised over the 3rd) and 2 > 1 > 3. We can notice that, in both cases, the 3rd person is the lowest in the hierarchy. Moreover, the difference between the two hierarchies has socio-cultural implications (in languages with a 1 > 2 > 3 hierarchy, the 1st person, or the speaker, is hierarchically superior, meaning that we can talk about a speaker-focused language, while the languages with 2 > 1 > 3 hierarchy are listener-oriented).\n\nLet us consider the above-mentioned Swahili examples and also imagine a language Y which has a 1 > 2 > 3 hierarchy. In the table below, T = tense, V = verb stem.\n\n| ‘I see yousg’ | ‘Yousg see me’\nSwahili | ni-na-wu-ona | wu-na-ni-ona\n| 1sg-T-2sg-V | 2sg-T-1sg-V\nY | 1sg-2sg-V-X | 1sg-2sg-V-X′\n\nIf we analyse the (artificial) examples from language Y, we notice, in this case, that the order of the two pronominal morphemes is identical (1st person appears before 2nd person, since it is hierarchically superior). The two examples do not differ in the order of two morphemes, but rather in the morpheme X (or X′), which shows that the hierarchy is or is not observed respectively. Thus, in the example ‘I see yousg’, morpheme X shows that the hierarchy is respected (i.e., that the person of the subject is hierarchically superior to that of the object), while in the sentence ‘yousg see me’, the morpheme X′ shows the opposite: the hierarchy is not followed, since the person of the object is hierarchically superior to the person of the subject.","source_file":"chapters/06-Verb.tex","source_line_start":858,"license":"CC-BY-4.0","retrieval_text":"Verb and verb phrase\nPronoun hierarchy\nverb morphology and argument structure\nIn many languages that we are familiar with, the subject and object are distinguished solely based on the word or morpheme order. For example, the difference between The dog sees the cat and The cat sees the dog is strictly due to the word order. Since the subject is, generally, the first in the sentence, we know that in the first sentence the dog is the subject and the cat is the object. Swapping the positions of the two, we also change their roles: so that in the second sentence, the cat becomes the subject, while the dog is the object.\n\nEven in the problems we have solved up to now, the same rules have applied. If we look back at Problem 6.1 (Swahili), we remember that the subject was always placed first, while the object was placed third. Thus, the structure ninawuona (ni-na-wu-ona = 1sg-present-2sg-‘see’) was translated as ‘I see yousg’, but, if we reverse the order of morphemes I and III (wunaniona), we get ‘yousg see me’.\n\nNevertheless, some languages work quite differently. The order of the two pronoun morphemes can be completely independent of their roles (subject or object), but rather depends on a predefined hierarchy of the persons together with another morpheme which shows whether the subject and object follow that hierarchy or not. The most common pronominal hierarchies are 1 > 2 > 3 (meaning that the 1st person is prioritised over the 2nd, which is then prioritised over the 3rd) and 2 > 1 > 3. We can notice that, in both cases, the 3rd person is the lowest in the hierarchy. Moreover, the difference between the two hierarchies has socio-cultural implications (in languages with a 1 > 2 > 3 hierarchy, the 1st person, or the speaker, is hierarchically superior, meaning that we can talk about a speaker-focused language, while the languages with 2 > 1 > 3 hierarchy are listener-oriented).\n\nLet us consider the above-mentioned Swahili examples and also imagine a language Y which has a 1 > 2 > 3 hierarchy. In the table below, T = tense, V = verb stem.\n\n| ‘I see yousg’ | ‘Yousg see me’\nSwahili | ni-na-wu-ona | wu-na-ni-ona\n| 1sg-T-2sg-V | 2sg-T-1sg-V\nY | 1sg-2sg-V-X | 1sg-2sg-V-X′\n\nIf we analyse the (artificial) examples from language Y, we notice, in this case, that the order of the two pronominal morphemes is identical (1st person appears before 2nd person, since it is hierarchically superior). The two examples do not differ in the order of two morphemes, but rather in the morpheme X (or X′), which shows that the hierarchy is or is not observed respectively. Thus, in the example ‘I see yousg’, morpheme X shows that the hierarchy is respected (i.e., that the person of the subject is hierarchically superior to that of the object), while in the sentence ‘yousg see me’, the morpheme X′ shows the opposite: the hierarchy is not followed, since the person of the object is hierarchically superior to the person of the subject."}
{"id":"book_method_c06_s08_p01","kind":"method","chapter":6,"chapter_title":"Verb and verb phrase","section":8,"section_title":"Verb semantics","topic":"verb morphology and argument structure","text":"Different English verbs can have the same stem in some languages, using some additional morphemes to alter the meaning. Let us remember from Problem 6.2 that the pairs ‘to argue’ – ‘to curse’ and ‘to beat’ – ‘to fight’ use the same verb stem in Dabida, although they are completely different in English. In this case, the meaning depends on the transitivity of the verb. The most common three types of morphemes which affect the verb semantics are:\n\n- Intensifiers: show that an action is performed “excessively” or “a lot” (compare, for example, to eat – to devour, or to hurt – to kill).\n\n- Mitigators: are the opposite of intensifiers and show that the intensity of an action is diminished (compare, for example, to eat – to taste or to see – to glance).\n\n- Causative marker: would be translated into English as to make someone do…. For example, to eat can become to feed (‘to make someone eat’) or to learn ‘to make someone learn’ = to teach).\n\nMoreover, certain languages can display a dichotomy (a contrast) between stative and active verbs. Active (or dynamic) verbs involve movement, an action which is performed (e.g., to hit, to kill, to look, to say, to eat, etc.), while stative verbs show a state or a feeling (to love, to hate, to like, to be good, to be ill, etc.). Certain languages treat these two verbal categories quite differently, using different TAM markers or even morpheme order.\n\nOther semantic considerations involve the notion of agent (A) and patient (P). These two terms are complementary to the pair subject-object. The agent is defined as the argument which performs the action, while the patient is that upon whom it is acted. Although these definitions may seem similar to those of the subject and object, these two arguments (agent and patient) are purely semantic. Thus, in the sentence The cat eats the fish, the subject is the cat, while the object is the fish. Moreover, the cat is also the agent (because it performs the action), while the fish is the patient (since it undergoes the action). In this case, therefore, S = A and O = P.\n\nOn the other hand, in the sentence The fish is eaten by the cat, the (grammatical) subject is the fish (since it displays concord with the verb), while the (prepositional) object is the cat. Nevertheless, from a purely semantic point of view, the agent is still the cat (it is still the cat that performs the action), while the fish is the patient.\n\nThus, we can consider the subject and object as syntactic arguments (which depend strictly on the sentence structure), while agent and patient are semantic arguments (they do not depend on the sentence structure, but strictly on the meaning of it).\n\nSome languages can mark the argument of an intransitive verb either as a patient or as an agent. The choice is made by the speaker based on semantic considerations. The agentive marking (i.e., the argument of the intransitive verb is marked as agent) is used to show that the action is undertaken by the subject, while the patientive marking is used to show that the action is rather undergone by the subject.\n\nFor example, in English this patientive marking is usually implied by using the structures to get X (e.g., to get lost, to get drowned, etc.). Another example is verbs that indicate accidental actions, for which a patientive marking would probably be favoured (to fall, to trip, etc.).\n\nThis type of marking is usually referred to as fluid-S alignment (which is a type of morphosyntactic alignment, as shown in the next chapter). There are two subtypes of fluid-S alignment:\n\n- Agentive-default in which the “default” (neutral) way of marking the intransitive argument is as agent. In this case, the patientive marking is used to highlight a lack of volition / control.","source_file":"chapters/06-Verb.tex","source_line_start":1056,"license":"CC-BY-4.0","retrieval_text":"Verb and verb phrase\nVerb semantics\nverb morphology and argument structure\nDifferent English verbs can have the same stem in some languages, using some additional morphemes to alter the meaning. Let us remember from Problem 6.2 that the pairs ‘to argue’ – ‘to curse’ and ‘to beat’ – ‘to fight’ use the same verb stem in Dabida, although they are completely different in English. In this case, the meaning depends on the transitivity of the verb. The most common three types of morphemes which affect the verb semantics are:\n\n- Intensifiers: show that an action is performed “excessively” or “a lot” (compare, for example, to eat – to devour, or to hurt – to kill).\n\n- Mitigators: are the opposite of intensifiers and show that the intensity of an action is diminished (compare, for example, to eat – to taste or to see – to glance).\n\n- Causative marker: would be translated into English as to make someone do…. For example, to eat can become to feed (‘to make someone eat’) or to learn ‘to make someone learn’ = to teach).\n\nMoreover, certain languages can display a dichotomy (a contrast) between stative and active verbs. Active (or dynamic) verbs involve movement, an action which is performed (e.g., to hit, to kill, to look, to say, to eat, etc.), while stative verbs show a state or a feeling (to love, to hate, to like, to be good, to be ill, etc.). Certain languages treat these two verbal categories quite differently, using different TAM markers or even morpheme order.\n\nOther semantic considerations involve the notion of agent (A) and patient (P). These two terms are complementary to the pair subject-object. The agent is defined as the argument which performs the action, while the patient is that upon whom it is acted. Although these definitions may seem similar to those of the subject and object, these two arguments (agent and patient) are purely semantic. Thus, in the sentence The cat eats the fish, the subject is the cat, while the object is the fish. Moreover, the cat is also the agent (because it performs the action), while the fish is the patient (since it undergoes the action). In this case, therefore, S = A and O = P.\n\nOn the other hand, in the sentence The fish is eaten by the cat, the (grammatical) subject is the fish (since it displays concord with the verb), while the (prepositional) object is the cat. Nevertheless, from a purely semantic point of view, the agent is still the cat (it is still the cat that performs the action), while the fish is the patient.\n\nThus, we can consider the subject and object as syntactic arguments (which depend strictly on the sentence structure), while agent and patient are semantic arguments (they do not depend on the sentence structure, but strictly on the meaning of it).\n\nSome languages can mark the argument of an intransitive verb either as a patient or as an agent. The choice is made by the speaker based on semantic considerations. The agentive marking (i.e., the argument of the intransitive verb is marked as agent) is used to show that the action is undertaken by the subject, while the patientive marking is used to show that the action is rather undergone by the subject.\n\nFor example, in English this patientive marking is usually implied by using the structures to get X (e.g., to get lost, to get drowned, etc.). Another example is verbs that indicate accidental actions, for which a patientive marking would probably be favoured (to fall, to trip, etc.).\n\nThis type of marking is usually referred to as fluid-S alignment (which is a type of morphosyntactic alignment, as shown in the next chapter). There are two subtypes of fluid-S alignment:\n\n- Agentive-default in which the “default” (neutral) way of marking the intransitive argument is as agent. In this case, the patientive marking is used to highlight a lack of volition / control."}
{"id":"book_method_c06_s08_p02","kind":"method","chapter":6,"chapter_title":"Verb and verb phrase","section":8,"section_title":"Verb semantics","topic":"verb morphology and argument structure","text":"- Agentive-default in which the “default” (neutral) way of marking the intransitive argument is as agent. In this case, the patientive marking is used to highlight a lack of volition / control.\n\n- Patientive-default on the other hand, uses patient marking as default. By using the agentive marking, the speaker highlights a certain degree of volition / control.","source_file":"chapters/06-Verb.tex","source_line_start":1056,"license":"CC-BY-4.0","retrieval_text":"Verb and verb phrase\nVerb semantics\nverb morphology and argument structure\n- Agentive-default in which the “default” (neutral) way of marking the intransitive argument is as agent. In this case, the patientive marking is used to highlight a lack of volition / control.\n\n- Patientive-default on the other hand, uses patient marking as default. By using the agentive marking, the speaker highlights a certain degree of volition / control."}
{"id":"book_method_c07_s01_p01","kind":"method","chapter":7,"chapter_title":"Syntax","section":1,"section_title":"Introduction","topic":"syntax, word order, focus, and alignment","text":"Syntax is concerned with the study of sentences and phrases. We remember that sentences can typically be considered to be formed from a noun phrase and a verb phrase (sentence = NP + VP)The abbreviations used are: NP = noun phrase, VP = verb phrase. (though not necessarily in that order). Therefore, syntax problems are nothing more than a noun morphology problem combined with a verb morphology problem.\n\nNevertheless, since NP and VP are related in the same structure, there can be some additional interactions between the two. For example, the noun phrase can contain morphemes indicating the roles of nouns in the sentence (subject, object, agent, patient, etc.). Since the arguments can be expressed though nouns (not only pronouns, as was the case in the previous chapter), there can also be a wider variety of distinctions in the verb phrase. For verb phrase problems, the principal distinctions were person (1, 2, 3), number (singular, dual, plural) and, sometimes, gender (masculine, feminine), but for syntax problems we can also add distinctions such as human (difference between nouns which refer to humans versus the rest).","source_file":"chapters/07-Syntax.tex","source_line_start":5,"license":"CC-BY-4.0","retrieval_text":"Syntax\nIntroduction\nsyntax, word order, focus, and alignment\nSyntax is concerned with the study of sentences and phrases. We remember that sentences can typically be considered to be formed from a noun phrase and a verb phrase (sentence = NP + VP)The abbreviations used are: NP = noun phrase, VP = verb phrase. (though not necessarily in that order). Therefore, syntax problems are nothing more than a noun morphology problem combined with a verb morphology problem.\n\nNevertheless, since NP and VP are related in the same structure, there can be some additional interactions between the two. For example, the noun phrase can contain morphemes indicating the roles of nouns in the sentence (subject, object, agent, patient, etc.). Since the arguments can be expressed though nouns (not only pronouns, as was the case in the previous chapter), there can also be a wider variety of distinctions in the verb phrase. For verb phrase problems, the principal distinctions were person (1, 2, 3), number (singular, dual, plural) and, sometimes, gender (masculine, feminine), but for syntax problems we can also add distinctions such as human (difference between nouns which refer to humans versus the rest)."}
{"id":"book_method_c07_s02_p01","kind":"method","chapter":7,"chapter_title":"Syntax","section":2,"section_title":"Word order","topic":"syntax, word order, focus, and alignment","text":"Word order is one of the most important phenomena in syntax problems. While in the NP and VP problems the number of words is relatively small, in sentences and phrases the number of words can increase considerably; therefore the order in which these words are placed plays a very important role.\n\nGenerally, when talking about the word order of a language, we refer to the order of the subject, verb, and object; therefore, we can have six possible patterns: SOV, SVO, VSO, VOS, OSV, OVS. Cross-linguistically, the SOV and SVO patterns are the most common; they account for patterns in around 90% of the world's languages. The next one is VSO, which occurs in approx. 8% of languages (such as Classical Arabic, Tagalog, or Celtic languages like Irish, Welsh, Breton, and Manx). The least common patterns are VOS, OVS and OSV, the latter being encountered in fewer than 0.5% of languages.Based on the data from https://wals.info/chapter/81. If we look closely, we notice that the most common patterns (SOV, SVO, VSO) are those in which the subject is placed before the object, while the languages in which the subject is placed after the verb account for under 5% of the languages. Thus, it is statistically probable that in a problem the subject is placed before the object.\n\nMoreover, it is important to mention the dichotomy between languages with fixed or rigid word order and those with a flexible word order.\n\nFixed word order can be directly assigned to one of the six patterns above, i.e., when a language has fixed word order, (almost) every sentence follows the same word order. All the other word orders are either ungrammatical or rarely used. Languages with fixed word order often have a simpler (inflectional) morphology in subjects and objects since the role of the arguments is well determined by their position in the sentence. Thus, if we consider the English examples The cat saw the boy and The boy saw the cat, we notice that both sentences have the same structure and the only difference is the word order. Nevertheless, there is no ambiguity between subject and object (even though they both appear morphologically identical – neither the subject nor the object is marked), since the fixed word order dictates that the subject is first (before the verb) and the object is placed after the verb.\n\nFlexible (variable) word order, on the other hand, cannot be directly assigned to one of the six patterns above, i.e., when a language has flexible word order, not all sentences have the same word order. All six patterns (or most of them) are grammatical and commonly used. Nevertheless, there can be a dominant word order, i.e., one word order that is favoured. Unlike the languages with fixed word order, those with variable word order tend to have a much richer inflectional system. This is due to the fact that, since word order is flexible, each argument needs to be morphologically marked in order to avoid any ambiguities. Let us consider the following three examples from Romanian:\n\nUriașul îl dă pe copil tatălui.\n\nTatălui îl dă uriașul pe copil.\n\nPe copil îl dă uriașul tatălui.\n\nAll of these sentences have the same meaning (‘The giant gives the child to the father.’), but the order of the arguments (subject, direct object, indirect object) differs in each case. Nevertheless, the meaning is clear and the role of each noun is well defined due to the morphological marking: the subject (uriașul) is unmarked, the direct object (copil) is preceded by pe, while the indirect object (tatălui) ends in the suffix -ui. In this instance, the suffix -ui is an example of the dative case, marking the indirect object.\n\nDue to this morphological marking, the flexible word order does not create ambiguities in sentences such as:\n\nTatăl îl dă pe copil uriașului. = ‘The father gives the child to the giant.’","source_file":"chapters/07-Syntax.tex","source_line_start":11,"license":"CC-BY-4.0","retrieval_text":"Syntax\nWord order\nsyntax, word order, focus, and alignment\nWord order is one of the most important phenomena in syntax problems. While in the NP and VP problems the number of words is relatively small, in sentences and phrases the number of words can increase considerably; therefore the order in which these words are placed plays a very important role.\n\nGenerally, when talking about the word order of a language, we refer to the order of the subject, verb, and object; therefore, we can have six possible patterns: SOV, SVO, VSO, VOS, OSV, OVS. Cross-linguistically, the SOV and SVO patterns are the most common; they account for patterns in around 90% of the world's languages. The next one is VSO, which occurs in approx. 8% of languages (such as Classical Arabic, Tagalog, or Celtic languages like Irish, Welsh, Breton, and Manx). The least common patterns are VOS, OVS and OSV, the latter being encountered in fewer than 0.5% of languages.Based on the data from https://wals.info/chapter/81. If we look closely, we notice that the most common patterns (SOV, SVO, VSO) are those in which the subject is placed before the object, while the languages in which the subject is placed after the verb account for under 5% of the languages. Thus, it is statistically probable that in a problem the subject is placed before the object.\n\nMoreover, it is important to mention the dichotomy between languages with fixed or rigid word order and those with a flexible word order.\n\nFixed word order can be directly assigned to one of the six patterns above, i.e., when a language has fixed word order, (almost) every sentence follows the same word order. All the other word orders are either ungrammatical or rarely used. Languages with fixed word order often have a simpler (inflectional) morphology in subjects and objects since the role of the arguments is well determined by their position in the sentence. Thus, if we consider the English examples The cat saw the boy and The boy saw the cat, we notice that both sentences have the same structure and the only difference is the word order. Nevertheless, there is no ambiguity between subject and object (even though they both appear morphologically identical – neither the subject nor the object is marked), since the fixed word order dictates that the subject is first (before the verb) and the object is placed after the verb.\n\nFlexible (variable) word order, on the other hand, cannot be directly assigned to one of the six patterns above, i.e., when a language has flexible word order, not all sentences have the same word order. All six patterns (or most of them) are grammatical and commonly used. Nevertheless, there can be a dominant word order, i.e., one word order that is favoured. Unlike the languages with fixed word order, those with variable word order tend to have a much richer inflectional system. This is due to the fact that, since word order is flexible, each argument needs to be morphologically marked in order to avoid any ambiguities. Let us consider the following three examples from Romanian:\n\nUriașul îl dă pe copil tatălui.\n\nTatălui îl dă uriașul pe copil.\n\nPe copil îl dă uriașul tatălui.\n\nAll of these sentences have the same meaning (‘The giant gives the child to the father.’), but the order of the arguments (subject, direct object, indirect object) differs in each case. Nevertheless, the meaning is clear and the role of each noun is well defined due to the morphological marking: the subject (uriașul) is unmarked, the direct object (copil) is preceded by pe, while the indirect object (tatălui) ends in the suffix -ui. In this instance, the suffix -ui is an example of the dative case, marking the indirect object.\n\nDue to this morphological marking, the flexible word order does not create ambiguities in sentences such as:\n\nTatăl îl dă pe copil uriașului. = ‘The father gives the child to the giant.’"}
{"id":"book_method_c07_s02_p02","kind":"method","chapter":7,"chapter_title":"Syntax","section":2,"section_title":"Word order","topic":"syntax, word order, focus, and alignment","text":"Tatăl îl dă pe copil uriașului. = ‘The father gives the child to the giant.’\n\nTatălui îl dă copilul pe uriaș. = ‘The child gives the giant to the father.’\n\nPe tată îl dă copilului uriașul. = ‘The giant gives the father to the child.’\n\nTypologically speaking, the word order in a sentence is classified only based on subject, object, and verb. However, in writing the rules for linguistics problems, it is important to include all the components of the sentence. Thus, for the sentence The child writes a letter with the pen, it is not enough to write S V O, but rather we should write S V O Instr. (subject, verb, object, instrumental). Moreover, when describing the word order, we usually talk about the main constituents and not the internal structure of those constituents. This means we describe the relative positions of the subject, object, verb, instrumental, location, time, etc.\nAs for the internal structure of the constituents, this covers the order of the noun and its modifiers (adjectives, possessives, etc.), which we can write as Adj. – Noun, or, more generally, Modifier – Noun. The reason why it is preferable to separate the two is that modifiers can occur together with different constituents (subject, object, instrumental, etc.), as in the sentence The happy child writes a new letter with the black pen\n\nThus, if we wanted to combine the two structures above (S V O Instr. and Adj. – Noun), we would have to write:\n\nV [Adj – O] [Adj – Instr]\n\nSo we would have to show that the adjective can be placed before the noun in each constituent (subject, object, instrumental). Moreover, the adjective is part of the noun phrase, so its position is strictly relative to the head of the phrase (the noun), independent of its role in the sentence (subject, object, etc.). Interestingly, in many languages the word order for the sentence is mirrored in the word order for the NP, assuming the verb/noun is the head of the VP/NP respectively. So if the sentence word order is, say verb-final, the NP order will also be head-final.","source_file":"chapters/07-Syntax.tex","source_line_start":11,"license":"CC-BY-4.0","retrieval_text":"Syntax\nWord order\nsyntax, word order, focus, and alignment\nTatăl îl dă pe copil uriașului. = ‘The father gives the child to the giant.’\n\nTatălui îl dă copilul pe uriaș. = ‘The child gives the giant to the father.’\n\nPe tată îl dă copilului uriașul. = ‘The giant gives the father to the child.’\n\nTypologically speaking, the word order in a sentence is classified only based on subject, object, and verb. However, in writing the rules for linguistics problems, it is important to include all the components of the sentence. Thus, for the sentence The child writes a letter with the pen, it is not enough to write S V O, but rather we should write S V O Instr. (subject, verb, object, instrumental). Moreover, when describing the word order, we usually talk about the main constituents and not the internal structure of those constituents. This means we describe the relative positions of the subject, object, verb, instrumental, location, time, etc.\nAs for the internal structure of the constituents, this covers the order of the noun and its modifiers (adjectives, possessives, etc.), which we can write as Adj. – Noun, or, more generally, Modifier – Noun. The reason why it is preferable to separate the two is that modifiers can occur together with different constituents (subject, object, instrumental, etc.), as in the sentence The happy child writes a new letter with the black pen\n\nThus, if we wanted to combine the two structures above (S V O Instr. and Adj. – Noun), we would have to write:\n\nV [Adj – O] [Adj – Instr]\n\nSo we would have to show that the adjective can be placed before the noun in each constituent (subject, object, instrumental). Moreover, the adjective is part of the noun phrase, so its position is strictly relative to the head of the phrase (the noun), independent of its role in the sentence (subject, object, etc.). Interestingly, in many languages the word order for the sentence is mirrored in the word order for the NP, assuming the verb/noun is the head of the VP/NP respectively. So if the sentence word order is, say verb-final, the NP order will also be head-final."}
{"id":"book_method_c07_s03_p01","kind":"method","chapter":7,"chapter_title":"Syntax","section":3,"section_title":"Focusing","topic":"syntax, word order, focus, and alignment","text":"Focusing describes the syntactic process in which one part of the sentence is emphasized. In English, focusing can be done by intonation in speaking (compare: He hit the dog and He hit the dog In the first sentence, the emphasis is on the action of hitting, while in the latter the emphasis is on the object). Moreover, a common way to focus the object in English is by using a cleft sentence (e.g., I saw a cat vs. It is a cat that I saw).\n\nIn certain languages, focusing can be done by changing the word order (in most cases, this means that the focused part is moved nearer to the beginning of the sentence) or by using certain specific markers. For example, there might be definite or indefinite articles specific for the focused form or there can be specific morphemes which signal the fact that some words are focused. In Wolof,This phenomenon was featured in a problem by Vlad A. Neacșu (RoLO 2019). for example, there are four sets of pronouns: subject (1sg = man), object (1sg = ma), verb-focus (1sg = damay), and object-focus (1sg = laa). The first two types (subject and object) are used by default; the verb-focus pronoun is used for the subject if the verb is emphasized, while the object-focus form is used instead of the usual object pronoun, if it is emphasized. Moreover, if the verb or the object are focused, they are moved to the beginning of the sentence. Below the four forms for 1sg and 2sg are given:\n\n| S | O | V-focus | O-focus\n1sg | man | ma | damay | laa\n2sg | yow | la | dangay | nga\n\nLet us consider the sentence ‘I saw yousg’ (in Wolof, the corresponding verb is gisoon). We can have the following three cases:\n\n- Neutral sentence = no part of the sentence is focused. The word order is the default one, SOV.\n\nMan la gisoon. (‘I saw yousg.’)\n\n- Sentence with focused verb = the V-focus pronoun is used to replace the subject and it is placed at the beginning of the sentence (it can be considered a focus marker), being immediately followed by the verb.\n\nDamay gisoon la. (‘I saw yousg.’)\n\n- Sentence with focused object = object becomes the first in the sentence, the rest of the order is preserved.\n\nNga man gisoon. (‘I saw yousg.’)","source_file":"chapters/07-Syntax.tex","source_line_start":293,"license":"CC-BY-4.0","retrieval_text":"Syntax\nFocusing\nsyntax, word order, focus, and alignment\nFocusing describes the syntactic process in which one part of the sentence is emphasized. In English, focusing can be done by intonation in speaking (compare: He hit the dog and He hit the dog In the first sentence, the emphasis is on the action of hitting, while in the latter the emphasis is on the object). Moreover, a common way to focus the object in English is by using a cleft sentence (e.g., I saw a cat vs. It is a cat that I saw).\n\nIn certain languages, focusing can be done by changing the word order (in most cases, this means that the focused part is moved nearer to the beginning of the sentence) or by using certain specific markers. For example, there might be definite or indefinite articles specific for the focused form or there can be specific morphemes which signal the fact that some words are focused. In Wolof,This phenomenon was featured in a problem by Vlad A. Neacșu (RoLO 2019). for example, there are four sets of pronouns: subject (1sg = man), object (1sg = ma), verb-focus (1sg = damay), and object-focus (1sg = laa). The first two types (subject and object) are used by default; the verb-focus pronoun is used for the subject if the verb is emphasized, while the object-focus form is used instead of the usual object pronoun, if it is emphasized. Moreover, if the verb or the object are focused, they are moved to the beginning of the sentence. Below the four forms for 1sg and 2sg are given:\n\n| S | O | V-focus | O-focus\n1sg | man | ma | damay | laa\n2sg | yow | la | dangay | nga\n\nLet us consider the sentence ‘I saw yousg’ (in Wolof, the corresponding verb is gisoon). We can have the following three cases:\n\n- Neutral sentence = no part of the sentence is focused. The word order is the default one, SOV.\n\nMan la gisoon. (‘I saw yousg.’)\n\n- Sentence with focused verb = the V-focus pronoun is used to replace the subject and it is placed at the beginning of the sentence (it can be considered a focus marker), being immediately followed by the verb.\n\nDamay gisoon la. (‘I saw yousg.’)\n\n- Sentence with focused object = object becomes the first in the sentence, the rest of the order is preserved.\n\nNga man gisoon. (‘I saw yousg.’)"}
{"id":"book_method_c07_s04_p01","kind":"method","chapter":7,"chapter_title":"Syntax","section":4,"section_title":"Morphosyntactic alignment","topic":"syntax, word order, focus, and alignment","text":"This refers to the way in which three verbal arguments (subject of intransitive verb, subject of transitive verb, object) behave. For simplicity, in this chapter we will use the following notation: Subject of intransitive verb = Subject = S, Subject of transitive verb = Agent = A, Direct object = Object = O.\n\nIn order to better illustrate this concept, let us consider the following problem:","source_file":"chapters/07-Syntax.tex","source_line_start":323,"license":"CC-BY-4.0","retrieval_text":"Syntax\nMorphosyntactic alignment\nsyntax, word order, focus, and alignment\nThis refers to the way in which three verbal arguments (subject of intransitive verb, subject of transitive verb, object) behave. For simplicity, in this chapter we will use the following notation: Subject of intransitive verb = Subject = S, Subject of transitive verb = Agent = A, Direct object = Object = O.\n\nIn order to better illustrate this concept, let us consider the following problem:"}
{"id":"book_method_c07_s05_p01","kind":"method","chapter":7,"chapter_title":"Syntax","section":5,"section_title":"Split alignment","topic":"syntax, word order, focus, and alignment","text":"Certain languages can have two (or more) types of alignments, each of them appearing in a specific linguistic context. For example, Pashto has a split alignment: it follows a nominative-accusative alignment in the present tense, but an ergative-absolutive alignment in the past. We can compare the following examples:\n\nZe wlarrem.\nI went\n‘I went.’\nZe dzem.\nI go\n‘I go.’\nDai me woleed.\nhim I saw\n‘I saw him.’\nZe yay weenem.\nI him see\n‘I see him.’\nZe yay woleedelem.\nme he saw\n‘He saw me.’\nDai me weenee.\nhe me sees\n‘He sees me.’\n\nWe notice that 1sg can be expressed in two ways: ze and me. In the past tense sentences (first row), ze is used for subject and object, while me is used for agent. Thus, the past follows an ergative-absolutive alignment, with me being used as the ergative form of 1sg and ze the absolutive form. At the same time, in the present tense sentences, we notice that ze is used for S and A, while me is used for O. Thus, the present tense follows a nominative-accusative alignment, with ze being the nominative form of 1sg and me the accusative form.\n\nAnother situation in which we can talk about the coexistence of two different alignments in the same language (but which is not considered split alignment) is that in which, historically speaking, a language had a certain alignment but, in time, due to changes in morphophonology, it came to use the direct alignment (completely unmarked). This can be easily observed in English, where we talk about a nominative-accusative alignment of the pronouns (he – him, I – me), but about a direct alignment of the nouns.\n\nIn linguistics problems the ergative-absolutive alignment is commonly found (together with the nominative-accusative one), while the split alignment is usually highlighted between these two types of alignment. Although in Pashto the context of the two alignments depends solely on the tense, the distinction is motivated for other reasons too: person, discourse prominence of arguments (noun vs pronoun), etc. Every time we see both transitive and intransitive sentences in a linguistics problem, we need to take into account the possibility that the language has a different alignment. Although nominative-accusative languages are more common, especially in the West, and therefore more familiar, ergative languages represent about a quarter of all world languages, and are particularly found in less known language families, and so are proportionally more likely to occur in linguistics problems! Another feature of ergative languages is that they are almost all verb-initial or verb-final, almost never SVO.","source_file":"chapters/07-Syntax.tex","source_line_start":588,"license":"CC-BY-4.0","retrieval_text":"Syntax\nSplit alignment\nsyntax, word order, focus, and alignment\nCertain languages can have two (or more) types of alignments, each of them appearing in a specific linguistic context. For example, Pashto has a split alignment: it follows a nominative-accusative alignment in the present tense, but an ergative-absolutive alignment in the past. We can compare the following examples:\n\nZe wlarrem.\nI went\n‘I went.’\nZe dzem.\nI go\n‘I go.’\nDai me woleed.\nhim I saw\n‘I saw him.’\nZe yay weenem.\nI him see\n‘I see him.’\nZe yay woleedelem.\nme he saw\n‘He saw me.’\nDai me weenee.\nhe me sees\n‘He sees me.’\n\nWe notice that 1sg can be expressed in two ways: ze and me. In the past tense sentences (first row), ze is used for subject and object, while me is used for agent. Thus, the past follows an ergative-absolutive alignment, with me being used as the ergative form of 1sg and ze the absolutive form. At the same time, in the present tense sentences, we notice that ze is used for S and A, while me is used for O. Thus, the present tense follows a nominative-accusative alignment, with ze being the nominative form of 1sg and me the accusative form.\n\nAnother situation in which we can talk about the coexistence of two different alignments in the same language (but which is not considered split alignment) is that in which, historically speaking, a language had a certain alignment but, in time, due to changes in morphophonology, it came to use the direct alignment (completely unmarked). This can be easily observed in English, where we talk about a nominative-accusative alignment of the pronouns (he – him, I – me), but about a direct alignment of the nouns.\n\nIn linguistics problems the ergative-absolutive alignment is commonly found (together with the nominative-accusative one), while the split alignment is usually highlighted between these two types of alignment. Although in Pashto the context of the two alignments depends solely on the tense, the distinction is motivated for other reasons too: person, discourse prominence of arguments (noun vs pronoun), etc. Every time we see both transitive and intransitive sentences in a linguistics problem, we need to take into account the possibility that the language has a different alignment. Although nominative-accusative languages are more common, especially in the West, and therefore more familiar, ergative languages represent about a quarter of all world languages, and are particularly found in less known language families, and so are proportionally more likely to occur in linguistics problems! Another feature of ergative languages is that they are almost all verb-initial or verb-final, almost never SVO."}
{"id":"book_method_c08_s01_p01","kind":"method","chapter":8,"chapter_title":"Semantics","section":1,"section_title":"Introduction","topic":"semantics and graph-based matching","text":"Semantics is the subfield of linguistics concerned with the study of meaning. Thus, semantics problems are not based on discovering the way certain words change their form (phonetics or phonology), get inflected or derived (morphology) or on the way in which words combine into sentences (syntax). This type of problem is strictly focused on the meaning of words and on the way two (or multiple) words can be combined to form a new word with a different meaning (e.g., in English we have words like rainbow which comes from rain + bow, thus the bow/arc/bent shape (in the sky) caused by rain). So, in this case, the focus is not on the combination process (morphologically, X + Y XY), but rather on the meaning that it has. Generally, semantics problems are chaos-and-order problems (the corpus is given in random order) and the corpus consists of words (or combinations of two to three words) which do not necessarily share any morphological feature, but rather a semantic one (belong to the same semantic field).\n\nWe need to remember that words that designate organs or body parts (liver, heart, eye, etc.) are the most “dangerous” ones, in the sense that their combinations often transcend their semantic field, taking on not entirely expected meanings (one of their most common uses is to express emotions or feelings, which can be connected to certain organs). For example, in Cameroon pidgin,A pidgin is a mix of languages, a simplified way of communication, developed between two or more groups of people who do not share a common language. This way of communication is not spoken as a primary or native language. In this case, Cameroon pidgin is based on the English language. the word ‘generous’ is translated as open han (‘open hand’), ‘wickedness’ = blak hat (‘black heart’), ‘hatred’ = bat hat (‘bad heart’), ‘dizziness’ = blak ai (‘black eye’), ‘poverty’ = drai han (‘dry hand’), and so on.From a problem by Aleka Blackwell (NACLO 2014). Therefore, we should pay extra attention when we encounter body parts or organs together with emotions or feelings in a semantics problem.\n\nFurthermore, this type of problem requires a certain intuition to solve, since there is no absolute approach or solving method. Below we will present one solving method which can be used (the graph method).","source_file":"chapters/08-Semantics.tex","source_line_start":5,"license":"CC-BY-4.0","retrieval_text":"Semantics\nIntroduction\nsemantics and graph-based matching\nSemantics is the subfield of linguistics concerned with the study of meaning. Thus, semantics problems are not based on discovering the way certain words change their form (phonetics or phonology), get inflected or derived (morphology) or on the way in which words combine into sentences (syntax). This type of problem is strictly focused on the meaning of words and on the way two (or multiple) words can be combined to form a new word with a different meaning (e.g., in English we have words like rainbow which comes from rain + bow, thus the bow/arc/bent shape (in the sky) caused by rain). So, in this case, the focus is not on the combination process (morphologically, X + Y XY), but rather on the meaning that it has. Generally, semantics problems are chaos-and-order problems (the corpus is given in random order) and the corpus consists of words (or combinations of two to three words) which do not necessarily share any morphological feature, but rather a semantic one (belong to the same semantic field).\n\nWe need to remember that words that designate organs or body parts (liver, heart, eye, etc.) are the most “dangerous” ones, in the sense that their combinations often transcend their semantic field, taking on not entirely expected meanings (one of their most common uses is to express emotions or feelings, which can be connected to certain organs). For example, in Cameroon pidgin,A pidgin is a mix of languages, a simplified way of communication, developed between two or more groups of people who do not share a common language. This way of communication is not spoken as a primary or native language. In this case, Cameroon pidgin is based on the English language. the word ‘generous’ is translated as open han (‘open hand’), ‘wickedness’ = blak hat (‘black heart’), ‘hatred’ = bat hat (‘bad heart’), ‘dizziness’ = blak ai (‘black eye’), ‘poverty’ = drai han (‘dry hand’), and so on.From a problem by Aleka Blackwell (NACLO 2014). Therefore, we should pay extra attention when we encounter body parts or organs together with emotions or feelings in a semantics problem.\n\nFurthermore, this type of problem requires a certain intuition to solve, since there is no absolute approach or solving method. Below we will present one solving method which can be used (the graph method)."}
{"id":"book_method_c08_s02_p01","kind":"method","chapter":8,"chapter_title":"Semantics","section":2,"section_title":"Graph method","topic":"semantics and graph-based matching","text":"A graph is a combination of points (nodes) connected by lines. In this method, the base words will be represented by the nodes, while their combinations will be represented by the lines. So, if we want to represent the structures blak ai and blak hat from above, the graph would be something like the one presented in fig:CPE-short-graph.\n\n(ai) at (0,0) ai;\n(blak) at (2.5,0) blak;\n(hat) at (5,0) hat;\n(ai) – (blak) node[anchor=south,inner sep=3pt,midway] blak ai;\n(blak) – (hat) node[anchor=south,inner sep=3pt,midway] blak hat;\n\nCaption: Sample graph for the structures blak ai and blak hat.\n\nFrom this graph, we understand the following:\n\n- the word blak combines with ai to form blak ai. Moreover, the direction of the arrow (from blak to ai) shows the order in which the words combine (so the resulting phrase is blak ai and not ai blak);\n\n- similarly, blak combines with hat to form blak hat.\n\nOne thing we need to take into account is which of these words are actually given in the corpus. Based on our previous example, the dataset we analysed contain only the phrases blak ai and blak hat, so we need to distinguish the words that are given from those that are not (in this case, we marked them in bold and non-italic). In handwriting, it is easier if we underline or circle them.\n\nIn order to make things simpler, another thing we can do is not waste time writing the word combinations on top of the arrows, since the direction of the arrow already shows us the combination order (we can mark the fact that that word appears in our corpus through a horizontal line – as if we underline the phrase which we did not write anymore). Thus, the graph above (fig:CPE-short-graph) would become as shown in fig:CPE-graph-brief.\n\n(ai) at (0,0) ai;\n(blak) at (4,0) blak;\n(hat) at (8,0) hat;\n(ai) – (blak) node[anchor=south,inner sep=3pt,midway] ;\n(blak) – (hat) node[anchor=south,inner sep=3pt,midway] ;\n\nCaption: Simpler representation of the graph shown in fig:CPE-short-graph.\n\nHere, the line on top of the arrow shows us that the two compound words are given in the corpus.\n\nIn order to solve a linguistics problem using this method, we need to follow the next steps:\n\n- Step 1. Create a graph for all the words in the target language. For this step, it is preferred that we do not look at the English translations so we can focus solely on the word structure and not on the possible combinations of meaning.\n\n- Step 2. Create a partial graph for the English words. By partial, we mean that we do not necessarily need to include all the words. Of course, if we can include them all, it is even better, but sometimes we might not be completely certain how some words are connected with one another. Thus, it suffices that we construct a partial graph (it is important that this graph contains only combinations we can be sure of, not “likely” ones).\n\n- Step 3. Compare the partial graph of English words with the total graph of the words in the given language and see where they would match.\n\n- Step 4. Translate what we can and, knowing the shape of the graph, continue filling in the rest of the words.\n\n2(1) at (0,0) dyè;\n(2) at (2.75,0) ‘house’;\n(3) at (5.5,0) ‘top’;\n(4) at (2.75,-2) cɛ̀m;\n(5) at (-2.75,0) tyɛ̀n;\n(6) at (8.25,0) ‘garment’;\n(7) at (5.5,-2) ɲíg;\n(8) at (8.25,-2) wàŋ;\n(1) – (2) node[anchor=south,inner sep=3pt,midway] ;\n(3) – (2) node[anchor=south,inner sep=3pt,midway] ‘roof’;\n(2) – (4) node[anchor=south,inner sep=3pt,midway,right] ;\n(1) – (5) node[anchor=south,inner sep=3pt,midway] ;\n(6) – (3) node[anchor=south,inner sep=3pt,midway] ‘hat’;\n(7) – (8) node[anchor=south,inner sep=3pt,midway] ;\n\nCaption: Partially solved graph.\n\nMoreover, we know that one of the words cɛ̀m and dyè means ‘bottom’.","source_file":"chapters/08-Semantics.tex","source_line_start":13,"license":"CC-BY-4.0","retrieval_text":"Semantics\nGraph method\nsemantics and graph-based matching\nA graph is a combination of points (nodes) connected by lines. In this method, the base words will be represented by the nodes, while their combinations will be represented by the lines. So, if we want to represent the structures blak ai and blak hat from above, the graph would be something like the one presented in fig:CPE-short-graph.\n\n(ai) at (0,0) ai;\n(blak) at (2.5,0) blak;\n(hat) at (5,0) hat;\n(ai) – (blak) node[anchor=south,inner sep=3pt,midway] blak ai;\n(blak) – (hat) node[anchor=south,inner sep=3pt,midway] blak hat;\n\nCaption: Sample graph for the structures blak ai and blak hat.\n\nFrom this graph, we understand the following:\n\n- the word blak combines with ai to form blak ai. Moreover, the direction of the arrow (from blak to ai) shows the order in which the words combine (so the resulting phrase is blak ai and not ai blak);\n\n- similarly, blak combines with hat to form blak hat.\n\nOne thing we need to take into account is which of these words are actually given in the corpus. Based on our previous example, the dataset we analysed contain only the phrases blak ai and blak hat, so we need to distinguish the words that are given from those that are not (in this case, we marked them in bold and non-italic). In handwriting, it is easier if we underline or circle them.\n\nIn order to make things simpler, another thing we can do is not waste time writing the word combinations on top of the arrows, since the direction of the arrow already shows us the combination order (we can mark the fact that that word appears in our corpus through a horizontal line – as if we underline the phrase which we did not write anymore). Thus, the graph above (fig:CPE-short-graph) would become as shown in fig:CPE-graph-brief.\n\n(ai) at (0,0) ai;\n(blak) at (4,0) blak;\n(hat) at (8,0) hat;\n(ai) – (blak) node[anchor=south,inner sep=3pt,midway] ;\n(blak) – (hat) node[anchor=south,inner sep=3pt,midway] ;\n\nCaption: Simpler representation of the graph shown in fig:CPE-short-graph.\n\nHere, the line on top of the arrow shows us that the two compound words are given in the corpus.\n\nIn order to solve a linguistics problem using this method, we need to follow the next steps:\n\n- Step 1. Create a graph for all the words in the target language. For this step, it is preferred that we do not look at the English translations so we can focus solely on the word structure and not on the possible combinations of meaning.\n\n- Step 2. Create a partial graph for the English words. By partial, we mean that we do not necessarily need to include all the words. Of course, if we can include them all, it is even better, but sometimes we might not be completely certain how some words are connected with one another. Thus, it suffices that we construct a partial graph (it is important that this graph contains only combinations we can be sure of, not “likely” ones).\n\n- Step 3. Compare the partial graph of English words with the total graph of the words in the given language and see where they would match.\n\n- Step 4. Translate what we can and, knowing the shape of the graph, continue filling in the rest of the words.\n\n2(1) at (0,0) dyè;\n(2) at (2.75,0) ‘house’;\n(3) at (5.5,0) ‘top’;\n(4) at (2.75,-2) cɛ̀m;\n(5) at (-2.75,0) tyɛ̀n;\n(6) at (8.25,0) ‘garment’;\n(7) at (5.5,-2) ɲíg;\n(8) at (8.25,-2) wàŋ;\n(1) – (2) node[anchor=south,inner sep=3pt,midway] ;\n(3) – (2) node[anchor=south,inner sep=3pt,midway] ‘roof’;\n(2) – (4) node[anchor=south,inner sep=3pt,midway,right] ;\n(1) – (5) node[anchor=south,inner sep=3pt,midway] ;\n(6) – (3) node[anchor=south,inner sep=3pt,midway] ‘hat’;\n(7) – (8) node[anchor=south,inner sep=3pt,midway] ;\n\nCaption: Partially solved graph.\n\nMoreover, we know that one of the words cɛ̀m and dyè means ‘bottom’."}
{"id":"book_method_c08_s02_p02","kind":"method","chapter":8,"chapter_title":"Semantics","section":2,"section_title":"Graph method","topic":"semantics and graph-based matching","text":"Moreover, we know that one of the words cɛ̀m and dyè means ‘bottom’.\n\nThe remaining English words are: ‘floor’ (which we know represents ‘house + bottom’), ‘grain’, ‘eyeball’, ‘restaurant’, and ‘sole of foot’. We can already assume that ‘sole of foot’ is connected to ‘bottom’ (the bottom of the footlegbody). Thus, if cɛ̀m is ‘bottom’, we would not be able to connect it to ‘foot’, therefore dyè = ‘bottom’ and tyɛ̀n = ‘foot’. We can now modify the graph (see fig:Lango-step4).\n\n2(1) at (0,0) ‘bottom’;\n(2) at (2.75,0) ‘house’;\n(3) at (5.5,0) ‘top’;\n(4) at (2.75,-2) cɛ̀m;\n(5) at (-2.75,0) ‘foot’;\n(6) at (8.25,0) ‘garment’;\n(7) at (5.5,-2) ɲíg;\n(8) at (8.25,-2) wàŋ;\n(1) – (2) node[anchor=south,inner sep=3pt,midway] ‘floor’;\n(3) – (2) node[anchor=south,inner sep=3pt,midway] ‘roof’;\n(2) – (4) node[anchor=south,inner sep=3pt,midway,right] ;\n(1) – (5) node[anchor=south,inner sep=3pt,midway] ‘sole’;\n(6) – (3) node[anchor=south,inner sep=3pt,midway] ‘hat’;\n(7) – (8) node[anchor=south,inner sep=3pt,midway] ;\n\nCaption: Partially solved graph, including ‘bottom’ and ‘foot’.\n\nWe are left with three words: ‘restaurant’, ‘eyeball’, and ‘grain’. Moreover, we know that one of them needs to be connected to ‘house’ (‘house’ + cɛ̀m), while the other two must be derived from one another (ɲíg and ɲíg wàŋ). Among the English words, the one which is most closely related semantically with ‘house’ is ‘restaurant’ (‘restaurant’ = ‘house’ + ‘food’), while ‘eyeball’ can be derived from ‘grain’ as in ‘eyeball’ = ‘grain’ + ‘eye’ (the grain of the eye).\n\nThus, we can make all the correspondences:","source_file":"chapters/08-Semantics.tex","source_line_start":13,"license":"CC-BY-4.0","retrieval_text":"Semantics\nGraph method\nsemantics and graph-based matching\nMoreover, we know that one of the words cɛ̀m and dyè means ‘bottom’.\n\nThe remaining English words are: ‘floor’ (which we know represents ‘house + bottom’), ‘grain’, ‘eyeball’, ‘restaurant’, and ‘sole of foot’. We can already assume that ‘sole of foot’ is connected to ‘bottom’ (the bottom of the footlegbody). Thus, if cɛ̀m is ‘bottom’, we would not be able to connect it to ‘foot’, therefore dyè = ‘bottom’ and tyɛ̀n = ‘foot’. We can now modify the graph (see fig:Lango-step4).\n\n2(1) at (0,0) ‘bottom’;\n(2) at (2.75,0) ‘house’;\n(3) at (5.5,0) ‘top’;\n(4) at (2.75,-2) cɛ̀m;\n(5) at (-2.75,0) ‘foot’;\n(6) at (8.25,0) ‘garment’;\n(7) at (5.5,-2) ɲíg;\n(8) at (8.25,-2) wàŋ;\n(1) – (2) node[anchor=south,inner sep=3pt,midway] ‘floor’;\n(3) – (2) node[anchor=south,inner sep=3pt,midway] ‘roof’;\n(2) – (4) node[anchor=south,inner sep=3pt,midway,right] ;\n(1) – (5) node[anchor=south,inner sep=3pt,midway] ‘sole’;\n(6) – (3) node[anchor=south,inner sep=3pt,midway] ‘hat’;\n(7) – (8) node[anchor=south,inner sep=3pt,midway] ;\n\nCaption: Partially solved graph, including ‘bottom’ and ‘foot’.\n\nWe are left with three words: ‘restaurant’, ‘eyeball’, and ‘grain’. Moreover, we know that one of them needs to be connected to ‘house’ (‘house’ + cɛ̀m), while the other two must be derived from one another (ɲíg and ɲíg wàŋ). Among the English words, the one which is most closely related semantically with ‘house’ is ‘restaurant’ (‘restaurant’ = ‘house’ + ‘food’), while ‘eyeball’ can be derived from ‘grain’ as in ‘eyeball’ = ‘grain’ + ‘eye’ (the grain of the eye).\n\nThus, we can make all the correspondences:"}
{"id":"book_method_c08_s05_p01","kind":"method","chapter":8,"chapter_title":"Semantics","section":5,"section_title":"Explanations","topic":"semantics and graph-based matching","text":"When two words combine (A–B), there are some phonological changes occurring in both of them. Thus, the first word (the prefix) undergoes some alternations at the end. If it ends in y, it will become iN (where N is a nasal which assimilates to the place of articulation of the next consonant – m for p/b and n for d/k) – e.g., zafin-kitrokely and zafim-paladia.\n\nThe following word (B) undergoes an initial consonant mutation, depending on whether the prefix ends in a vowel or a consonant, as follows:\n\nFirst consonant of B | after vowel | after consonant\nl | l | d\nv | | b\nk | | k\n(?) | f | p\n\nSince we are asked to find (?) (in task c), the only rule we can deduce is that there is no initial consonant mutation if the word before ends in a vowel.\n\nInterestingly, the words for ‘great-grandson’, ‘great-great-grandson’, etc. are based on vertical position of body parts: the more distant the descendant, the lower the position of the body part used. The terms ‘great-great-grandson’, ‘great-great-great-grandson’, and ‘great-great-great-great-grandson’ are formed by compounding zafy = ‘grandson’ with ‘knee’, ‘ankle’, and ‘sole’, respectively.","source_file":"chapters/08-Semantics.tex","source_line_start":983,"license":"CC-BY-4.0","retrieval_text":"Semantics\nExplanations\nsemantics and graph-based matching\nWhen two words combine (A–B), there are some phonological changes occurring in both of them. Thus, the first word (the prefix) undergoes some alternations at the end. If it ends in y, it will become iN (where N is a nasal which assimilates to the place of articulation of the next consonant – m for p/b and n for d/k) – e.g., zafin-kitrokely and zafim-paladia.\n\nThe following word (B) undergoes an initial consonant mutation, depending on whether the prefix ends in a vowel or a consonant, as follows:\n\nFirst consonant of B | after vowel | after consonant\nl | l | d\nv | | b\nk | | k\n(?) | f | p\n\nSince we are asked to find (?) (in task c), the only rule we can deduce is that there is no initial consonant mutation if the word before ends in a vowel.\n\nInterestingly, the words for ‘great-grandson’, ‘great-great-grandson’, etc. are based on vertical position of body parts: the more distant the descendant, the lower the position of the body part used. The terms ‘great-great-grandson’, ‘great-great-great-grandson’, and ‘great-great-great-great-grandson’ are formed by compounding zafy = ‘grandson’ with ‘knee’, ‘ankle’, and ‘sole’, respectively."}
{"id":"book_method_c09_s01_p01","kind":"method","chapter":9,"chapter_title":"Number systems","section":1,"section_title":"Introduction","topic":"number systems","text":"In order to understand problems about number systems, we firstly need to understand the concept of base and the characteristics of such systems.\n\nIn Romanian, for example, the number 432 is written patru sute treizeci și doi. It can be segmented into morphemes as follows: patru (‘four’) sute (‘hundreds’) trei (‘three’)-zeci (‘tens’) și (‘and’) doi (‘two’), and each morpheme has a well defined role. The morphemes patru (‘four’), trei (‘three’), doi (‘two’) represent the digits, the basic units, while the morphemes sute (‘hundreds’) and zeci (‘tens’) represent the orders, i.e., some greater base words. Finally, the morpheme și (‘and’) denotes addition. Since in this case the orders are 10 and 100 (101, 102), we are talking about a base-10 (or decimal) system.\n\nMoreover, word order plays an important role, together with the way in which the multiplication and the addition are marked. Returning to the example patru sute treizeci și doi, we can deduce the following:\n\n- the general order is from big to small (we first write the hundreds, then the tens and finally the units);\n\n- the multiplier is placed before the order (thus, we write patru sute and not *sute patru);\n\n- the morpheme for ‘tens’ is written as part of the same word with its multiplier (treizeci, not *trei zeci);\n\n- the tens are separated from the units by the word și.\n\nThere is also a simpler method to show all of these rules, by writing the general structure of the number:\n\n100X + 10Y + Z = X sute Y-zeci și Z\n\nThis rule combines all four rules from above, showing the sequence of the orders (hundreds, tens, units), the position of the multiplier (as a separate word before the hundreds and combined with the ending zeci for tens), as well as the morpheme și added between tens and units.\n\nAlthough base 10 is the most common base and most languages use it, there are also other bases, among which the most common are base 6 and base 20. Less common, but occurring in a reasonable number of languages are bases 4, 5, 12, and 60, while bases 7, 9, 11, and 13 are extremely rare (at the moment, no known natural language uses them), so that, in linguistics problems, we can assume from the beginning that the system is highly unlikely to have the bases 7, 9, 11, 13.\n\nA special category of bases is 22–28, which are found in languages whose number system is based on body parts, such as Oksapmin or Kaugel.\n\nIn order to better understand the concept of base, let us consider how the numbers 10, 25, and 100 would be written in a base-6 system. Since the base is 6, the orders are 6 (61), 36 (62), 216 (63), and so on. Therefore, the number 10 will be written as 6+4 (or 4+6, depending on the word order), number 25 will be written as (46)+1, while 100 will be written as (236)+(46)+4.\n\nSo, in problems involving number systems, there are two important characteristics:\n\n- word order (or direction): big-to-small or small-to-big or a mixture;\n\n- the way in which addition and multiplication are marked (or whether they are not overtly marked).","source_file":"chapters/09-Numbers.tex","source_line_start":6,"license":"CC-BY-4.0","retrieval_text":"Number systems\nIntroduction\nnumber systems\nIn order to understand problems about number systems, we firstly need to understand the concept of base and the characteristics of such systems.\n\nIn Romanian, for example, the number 432 is written patru sute treizeci și doi. It can be segmented into morphemes as follows: patru (‘four’) sute (‘hundreds’) trei (‘three’)-zeci (‘tens’) și (‘and’) doi (‘two’), and each morpheme has a well defined role. The morphemes patru (‘four’), trei (‘three’), doi (‘two’) represent the digits, the basic units, while the morphemes sute (‘hundreds’) and zeci (‘tens’) represent the orders, i.e., some greater base words. Finally, the morpheme și (‘and’) denotes addition. Since in this case the orders are 10 and 100 (101, 102), we are talking about a base-10 (or decimal) system.\n\nMoreover, word order plays an important role, together with the way in which the multiplication and the addition are marked. Returning to the example patru sute treizeci și doi, we can deduce the following:\n\n- the general order is from big to small (we first write the hundreds, then the tens and finally the units);\n\n- the multiplier is placed before the order (thus, we write patru sute and not *sute patru);\n\n- the morpheme for ‘tens’ is written as part of the same word with its multiplier (treizeci, not *trei zeci);\n\n- the tens are separated from the units by the word și.\n\nThere is also a simpler method to show all of these rules, by writing the general structure of the number:\n\n100X + 10Y + Z = X sute Y-zeci și Z\n\nThis rule combines all four rules from above, showing the sequence of the orders (hundreds, tens, units), the position of the multiplier (as a separate word before the hundreds and combined with the ending zeci for tens), as well as the morpheme și added between tens and units.\n\nAlthough base 10 is the most common base and most languages use it, there are also other bases, among which the most common are base 6 and base 20. Less common, but occurring in a reasonable number of languages are bases 4, 5, 12, and 60, while bases 7, 9, 11, and 13 are extremely rare (at the moment, no known natural language uses them), so that, in linguistics problems, we can assume from the beginning that the system is highly unlikely to have the bases 7, 9, 11, 13.\n\nA special category of bases is 22–28, which are found in languages whose number system is based on body parts, such as Oksapmin or Kaugel.\n\nIn order to better understand the concept of base, let us consider how the numbers 10, 25, and 100 would be written in a base-6 system. Since the base is 6, the orders are 6 (61), 36 (62), 216 (63), and so on. Therefore, the number 10 will be written as 6+4 (or 4+6, depending on the word order), number 25 will be written as (46)+1, while 100 will be written as (236)+(46)+4.\n\nSo, in problems involving number systems, there are two important characteristics:\n\n- word order (or direction): big-to-small or small-to-big or a mixture;\n\n- the way in which addition and multiplication are marked (or whether they are not overtly marked)."}
{"id":"book_method_c09_s01_p02","kind":"method","chapter":9,"chapter_title":"Number systems","section":1,"section_title":"Introduction","topic":"number systems","text":"- the way in which addition and multiplication are marked (or whether they are not overtly marked).\n\nIn terms of (1), remember that it is not necessary that the word order is strictly ascending or descending, and there can also be exceptions. For example, in German, 432 is written vierhundertzweiunddreißig (vier-hundert-zwei-und-drei-ßig = 4-‘hundreds’-2-‘plus’-3-‘tens’), so the order is hundreds-units-tens. In terms of (2), note that in French, for example, in the number 2,510 (deux mille cinq cent dix = ‘two’ - ‘thousand’ - ‘five’ - ‘hundred’ - ‘ten’), both the addition and the multiplication are implicit and as such, if the digit is placed before the order, it gets multiplied, while addition occurs every time after the order (e.g., 2,510 can be written as (21000)+(5100)+10). Moreover, if the addition or multiplication is marked, this will be done using a rather pervasive morpheme, word, or group of words.\n\nFor this problem, notice that counting the digits and understanding the structure of the numbers is extremely important for number problems. Moreover, carryovers offer valuable information for problems in which equalities are given, and they can usually be used to infer the digits 2 and 3.\n\nMoreover, it is important to mention that, in order to be able to solve the problem, we need to use the fact that the problem is self-sufficient, otherwise we cannot deduce that the base is 5. Let us consider a base B and the following values for the digits: omme = 2, kwimane = 4, abba = B-4, umbea = B-2. The equalities become:\n\n- (B–2) + B + 4 = 2B + 2\n\n- 2B + 4 + (B + (B–4)) = 4B\n\nIn this case, the two equalities hold, no matter the base (e.g., for base 13 we would get the equalities: (1) 11 + 17 = 28 and (2) 30 + 22 = 52). Nevertheless, task (b) asks for the translation of 1, so abba needs to be 1 (we know that umbea is bigger than abba, since B-4 < B-2), and if abba = 1 = B-4 B = 5.\n\nThis is the complete thought process based upon which we deduce the base is 5. However, generally, we expect that the problem features all (or almost all) digits, enabling us to estimate the base based on counting the digits.","source_file":"chapters/09-Numbers.tex","source_line_start":6,"license":"CC-BY-4.0","retrieval_text":"Number systems\nIntroduction\nnumber systems\n- the way in which addition and multiplication are marked (or whether they are not overtly marked).\n\nIn terms of (1), remember that it is not necessary that the word order is strictly ascending or descending, and there can also be exceptions. For example, in German, 432 is written vierhundertzweiunddreißig (vier-hundert-zwei-und-drei-ßig = 4-‘hundreds’-2-‘plus’-3-‘tens’), so the order is hundreds-units-tens. In terms of (2), note that in French, for example, in the number 2,510 (deux mille cinq cent dix = ‘two’ - ‘thousand’ - ‘five’ - ‘hundred’ - ‘ten’), both the addition and the multiplication are implicit and as such, if the digit is placed before the order, it gets multiplied, while addition occurs every time after the order (e.g., 2,510 can be written as (21000)+(5100)+10). Moreover, if the addition or multiplication is marked, this will be done using a rather pervasive morpheme, word, or group of words.\n\nFor this problem, notice that counting the digits and understanding the structure of the numbers is extremely important for number problems. Moreover, carryovers offer valuable information for problems in which equalities are given, and they can usually be used to infer the digits 2 and 3.\n\nMoreover, it is important to mention that, in order to be able to solve the problem, we need to use the fact that the problem is self-sufficient, otherwise we cannot deduce that the base is 5. Let us consider a base B and the following values for the digits: omme = 2, kwimane = 4, abba = B-4, umbea = B-2. The equalities become:\n\n- (B–2) + B + 4 = 2B + 2\n\n- 2B + 4 + (B + (B–4)) = 4B\n\nIn this case, the two equalities hold, no matter the base (e.g., for base 13 we would get the equalities: (1) 11 + 17 = 28 and (2) 30 + 22 = 52). Nevertheless, task (b) asks for the translation of 1, so abba needs to be 1 (we know that umbea is bigger than abba, since B-4 < B-2), and if abba = 1 = B-4 B = 5.\n\nThis is the complete thought process based upon which we deduce the base is 5. However, generally, we expect that the problem features all (or almost all) digits, enabling us to estimate the base based on counting the digits."}
{"id":"book_method_c09_s02_p01","kind":"method","chapter":9,"chapter_title":"Number systems","section":2,"section_title":"Overcounting","topic":"number systems","text":"It is interesting to note the phenomenon in which we write both ngui X and ngui (X+1)-gonaga, although the second part is redundant. Basically, we can consider this structure to mean Z units after ngui X, towards ngui (X+1)-gonaga. For example, the number 49 (ngui tebo, ngui mane-gonaga maria) can be read as ‘four (units) away from the third group of 15s towards the fourth (group of 15s)’. This is another example of overcounting in which the units are enclosed between two consecutive multiples of the base.","source_file":"chapters/09-Numbers.tex","source_line_start":377,"license":"CC-BY-4.0","retrieval_text":"Number systems\nOvercounting\nnumber systems\nIt is interesting to note the phenomenon in which we write both ngui X and ngui (X+1)-gonaga, although the second part is redundant. Basically, we can consider this structure to mean Z units after ngui X, towards ngui (X+1)-gonaga. For example, the number 49 (ngui tebo, ngui mane-gonaga maria) can be read as ‘four (units) away from the third group of 15s towards the fourth (group of 15s)’. This is another example of overcounting in which the units are enclosed between two consecutive multiples of the base."}
{"id":"book_method_c09_s03_p01","kind":"method","chapter":9,"chapter_title":"Number systems","section":3,"section_title":"Subtractive systems","topic":"number systems","text":"We have seen so far that the basic operations in every number system are addition and multiplication. There is, however, a special type of number system, subtractive systems, in which, besides addition and multiplication, subtraction also plays an important role. For example, in Problem 9.4, the numbers were formed as M+1, M+2, M+3, N (where M and N are orders – we ignore the fact that M was derived from N). In subtractive systems, we can talk about numbers formed as {M, M+1, M+2, N-2, N-1, N} or {M, M+1, M+2, M+3, N-1, N}. Thus, 19 can, for example, be written as 20-1 instead of 16+3. It is important to not confuse subtractive systems with overcounting. In the case of overcounting, no subtraction is involved, but rather the order we use is one unit higher (since it is counted towards it), but still, the units are added. In subtractive systems, the units (or some of them) are subtracted from the order.\n\nCurrently, no purely subtractive system is known (in which addition does not play some role), and all subtractive systems use both addition and subtraction. Most subtractive systems only employ subtraction for 1 (therefore, for example, in such a base-10 subtractive system, the number 38 is written as 30 + 8, but the number 39 is written as 40-1).","source_file":"chapters/09-Numbers.tex","source_line_start":755,"license":"CC-BY-4.0","retrieval_text":"Number systems\nSubtractive systems\nnumber systems\nWe have seen so far that the basic operations in every number system are addition and multiplication. There is, however, a special type of number system, subtractive systems, in which, besides addition and multiplication, subtraction also plays an important role. For example, in Problem 9.4, the numbers were formed as M+1, M+2, M+3, N (where M and N are orders – we ignore the fact that M was derived from N). In subtractive systems, we can talk about numbers formed as {M, M+1, M+2, N-2, N-1, N} or {M, M+1, M+2, M+3, N-1, N}. Thus, 19 can, for example, be written as 20-1 instead of 16+3. It is important to not confuse subtractive systems with overcounting. In the case of overcounting, no subtraction is involved, but rather the order we use is one unit higher (since it is counted towards it), but still, the units are added. In subtractive systems, the units (or some of them) are subtracted from the order.\n\nCurrently, no purely subtractive system is known (in which addition does not play some role), and all subtractive systems use both addition and subtraction. Most subtractive systems only employ subtraction for 1 (therefore, for example, in such a base-10 subtractive system, the number 38 is written as 30 + 8, but the number 39 is written as 40-1)."}
{"id":"book_method_c09_s04_p01","kind":"method","chapter":9,"chapter_title":"Number systems","section":4,"section_title":"Body-part counting systems","topic":"number systems","text":"In this type of system, the names of the digits are derived from the names of body parts. Thus, 1 = ‘little finger’, 2 = ‘ring finger’, 3 = ‘middle finger’, 4 = ‘index finger’, 5 = ‘thumb’, 6 = ‘hand’, 7 = ‘elbow’, 8 = ‘arm’, 9 = ‘shoulder’, 10 = ‘ear’, 11 = ‘head’. Depending on the language, the number of body parts which are used can vary (some languages also include terms for ‘wrist’, ‘forearm’, ‘arm’, ‘eye’, etc.). We mentioned in the beginning of this chapter that the base of these systems is usually between 22 and 28. The interesting thing about these systems starts as soon as we reach the word for ‘head’ (or ‘forehead’, ‘nose’, etc.), which represents the centre. From here on, the words are repeated in reverse order, with an additional morpheme/word which means ‘opposite/other’. Thus, continuing the series above, 12 = ‘opposite ear’, 13 = ‘opposite shoulder’, 14 = ‘opposite arm’, 15 = ‘opposite elbow’, 16 = ‘opposite hand’, 17 = ‘opposite thumb’, and so on.\n\nAt first sight, this particularity, although interesting, does not seem to pose any problems, but let us consider the following examples:\nLet us first consider a classic number system (not based on body parts) – for example, Japanese – and let us consider two pairs of numbers:\n\nsan – juusan and go – juugo\nIn the case of these numbers, we have pairs following the pattern X – juu-X. So, if we are told that san = 3 and juusan = 13, we could instantly deduce that juu- = 10 (since 13 - 3 = 10), and, if, additionally, we are told that go = 5, we will immediately claim that juugo = 15.\n\nLet us now consider a body-part-based system – e.g., the Kombai system, used in New Guinea – and let us analyse the following numbers:\n\nworo = 4, imofo woro = 20, go = 6, imofo go = ?\nIf we followed the same method as in the previous case, we would again notice the pattern X – imofo X, and based on the pair 4 – 20, we would deduce that imofo = 16, so we would say that imofo go = 22. Nevertheless, in reality, imofo go = 18, since the Kombai number system is based on body parts, centred on 12. Therefore, we notice that for these systems, the order is not obtained by subtracting two numbers like X and imofo X, but rather by adding them.\n\nFor Japanese: juusan - san = 10 = juugo - go, but for Kombai woro + imofo woro = go + imofo go. This is the main issue (and difficulty) with this type of system: the fact that it can easily be confused with a classic system, but, in fact, the base is obtained by adding two similar numbers rather than subtracting them.\n\nA separate discussion concerns what one considers to be the base of such a system. We return to the example of Kombai, where we mentioned that the system is ``centred on 12'', but imofo go + go = 24. For linguistics problems, it is preferred that the base is considered as the double of the centre, because, in this way, we will have the relation imofo X = 24 - X. If we considered the base to be 12, the relation between the two numbers would become: X = 12 - a, while imofo X = 12 + a, which is much harder to use in practice.","source_file":"chapters/09-Numbers.tex","source_line_start":864,"license":"CC-BY-4.0","retrieval_text":"Number systems\nBody-part counting systems\nnumber systems\nIn this type of system, the names of the digits are derived from the names of body parts. Thus, 1 = ‘little finger’, 2 = ‘ring finger’, 3 = ‘middle finger’, 4 = ‘index finger’, 5 = ‘thumb’, 6 = ‘hand’, 7 = ‘elbow’, 8 = ‘arm’, 9 = ‘shoulder’, 10 = ‘ear’, 11 = ‘head’. Depending on the language, the number of body parts which are used can vary (some languages also include terms for ‘wrist’, ‘forearm’, ‘arm’, ‘eye’, etc.). We mentioned in the beginning of this chapter that the base of these systems is usually between 22 and 28. The interesting thing about these systems starts as soon as we reach the word for ‘head’ (or ‘forehead’, ‘nose’, etc.), which represents the centre. From here on, the words are repeated in reverse order, with an additional morpheme/word which means ‘opposite/other’. Thus, continuing the series above, 12 = ‘opposite ear’, 13 = ‘opposite shoulder’, 14 = ‘opposite arm’, 15 = ‘opposite elbow’, 16 = ‘opposite hand’, 17 = ‘opposite thumb’, and so on.\n\nAt first sight, this particularity, although interesting, does not seem to pose any problems, but let us consider the following examples:\nLet us first consider a classic number system (not based on body parts) – for example, Japanese – and let us consider two pairs of numbers:\n\nsan – juusan and go – juugo\nIn the case of these numbers, we have pairs following the pattern X – juu-X. So, if we are told that san = 3 and juusan = 13, we could instantly deduce that juu- = 10 (since 13 - 3 = 10), and, if, additionally, we are told that go = 5, we will immediately claim that juugo = 15.\n\nLet us now consider a body-part-based system – e.g., the Kombai system, used in New Guinea – and let us analyse the following numbers:\n\nworo = 4, imofo woro = 20, go = 6, imofo go = ?\nIf we followed the same method as in the previous case, we would again notice the pattern X – imofo X, and based on the pair 4 – 20, we would deduce that imofo = 16, so we would say that imofo go = 22. Nevertheless, in reality, imofo go = 18, since the Kombai number system is based on body parts, centred on 12. Therefore, we notice that for these systems, the order is not obtained by subtracting two numbers like X and imofo X, but rather by adding them.\n\nFor Japanese: juusan - san = 10 = juugo - go, but for Kombai woro + imofo woro = go + imofo go. This is the main issue (and difficulty) with this type of system: the fact that it can easily be confused with a classic system, but, in fact, the base is obtained by adding two similar numbers rather than subtracting them.\n\nA separate discussion concerns what one considers to be the base of such a system. We return to the example of Kombai, where we mentioned that the system is ``centred on 12'', but imofo go + go = 24. For linguistics problems, it is preferred that the base is considered as the double of the centre, because, in this way, we will have the relation imofo X = 24 - X. If we considered the base to be 12, the relation between the two numbers would become: X = 12 - a, while imofo X = 12 + a, which is much harder to use in practice."}
{"id":"book_method_c09_s05_p01","kind":"method","chapter":9,"chapter_title":"Number systems","section":5,"section_title":"Time","topic":"number systems","text":"Problems related to time (reading the clock, calendar, etc.) can be considered a subtype of number systems since these problems will almost always feature numbers. It is rather common for the names of the days of the week and the months of the year to be formed based on numbers (‘Monday’ = day 1, ‘March’ = month 3). This phenomenon is obvious in Chinese, where, starting from the numbers 1 = 一, 2 = 二, 3 = 三, 4 = 四, we can write the days of the week and the months as follows: ‘Monday’ = 星期一, ‘Tuesday’ = 星期二, ‘Wednesday’ = 星期三, ‘Thursday’ = 星期四; ‘January’ = 一月, ‘February’ = 二月, ‘March’ = 三月, ‘April’ = 四月.\n\nAlthough these problems are relatively rare at the level of international linguistics competitions, they can feature some interesting phenomena, as shown in the following two problems.\n\nIn reality, the Swahili system is much more interesting and simpler than it may seem. In countries where Swahili is spoken, the sun rises at around 7 AM so the people consider 7 AM to be the first hour of the day (8 AM is the second hour, etc.), and the sunrise is also the moment in which the day changes. The words usiku and asubuhi also refer strictly to sunrise and sunset (usiku = between 7 PM and 6.59 AM = ‘after sunset’, and asubuhi = between 7 AM and 6.59 PM = ‘after sunrise’).\n\nThus, the days of the week are named: day 1, day 2 etc. (first day is Saturday) and the table becomes:\n\nHour | Day of the week\n1 | moja | jumamosi | ‘Saturday’\n2 | mbili | jumapili | ‘Sunday’\n3 | tatu | | ‘Monday’\n4 | nne | jumanne | ‘Tuesday’\n5 | tano | | ‘Wednesday’\n6 | sita | | ‘Thursday’\n7 | saba | | ‘Friday’\nsaa X = (X+6) ‘o'clock’\n\nThis system of counting hours based on the sunrise is rather common in Equatorial Africa, where the day length is relatively constant, and a similar system is also used in Ethiopia. What is important to remember is that every language is used to express a culture and that culture might be completely different from yours. Therefore, when approaching a linguistics problem, it is important to keep an open mind and not expect all things to work in the way you are used to (e.g., the day changes at midnight, the numbers are in base 10, the week has 7 days, etc.).","source_file":"chapters/09-Numbers.tex","source_line_start":887,"license":"CC-BY-4.0","retrieval_text":"Number systems\nTime\nnumber systems\nProblems related to time (reading the clock, calendar, etc.) can be considered a subtype of number systems since these problems will almost always feature numbers. It is rather common for the names of the days of the week and the months of the year to be formed based on numbers (‘Monday’ = day 1, ‘March’ = month 3). This phenomenon is obvious in Chinese, where, starting from the numbers 1 = 一, 2 = 二, 3 = 三, 4 = 四, we can write the days of the week and the months as follows: ‘Monday’ = 星期一, ‘Tuesday’ = 星期二, ‘Wednesday’ = 星期三, ‘Thursday’ = 星期四; ‘January’ = 一月, ‘February’ = 二月, ‘March’ = 三月, ‘April’ = 四月.\n\nAlthough these problems are relatively rare at the level of international linguistics competitions, they can feature some interesting phenomena, as shown in the following two problems.\n\nIn reality, the Swahili system is much more interesting and simpler than it may seem. In countries where Swahili is spoken, the sun rises at around 7 AM so the people consider 7 AM to be the first hour of the day (8 AM is the second hour, etc.), and the sunrise is also the moment in which the day changes. The words usiku and asubuhi also refer strictly to sunrise and sunset (usiku = between 7 PM and 6.59 AM = ‘after sunset’, and asubuhi = between 7 AM and 6.59 PM = ‘after sunrise’).\n\nThus, the days of the week are named: day 1, day 2 etc. (first day is Saturday) and the table becomes:\n\nHour | Day of the week\n1 | moja | jumamosi | ‘Saturday’\n2 | mbili | jumapili | ‘Sunday’\n3 | tatu | | ‘Monday’\n4 | nne | jumanne | ‘Tuesday’\n5 | tano | | ‘Wednesday’\n6 | sita | | ‘Thursday’\n7 | saba | | ‘Friday’\nsaa X = (X+6) ‘o'clock’\n\nThis system of counting hours based on the sunrise is rather common in Equatorial Africa, where the day length is relatively constant, and a similar system is also used in Ethiopia. What is important to remember is that every language is used to express a culture and that culture might be completely different from yours. Therefore, when approaching a linguistics problem, it is important to keep an open mind and not expect all things to work in the way you are used to (e.g., the day changes at midnight, the numbers are in base 10, the week has 7 days, etc.)."}
{"id":"book_method_c10_s01_p01","kind":"method","chapter":10,"chapter_title":"Other types of problems","section":1,"section_title":"Problems based on orientation systems","topic":"orientation, kinship, and other structural problems","text":"The purpose of orientation system problems is to identify the way in which a specific language expresses directions (relative positions of objects, such as in ‘in front’, ‘behind’, ‘to the left’, ‘to the right’, etc. or directions towards something: ‘go ahead’, ‘turn left’, ‘turn right’, etc.). These problems are typically easy to recognise since they usually contain the image of a map or a similar diagram or picture.\n\nTypologically speaking, orientation systems can be classified into two categories: absolute or relative referential systems. In absolute referential systems, the directions are relative to one or more fixed points (e.g., cardinal directions or geographical locations). Probably the best-known language which uses an absolute orientation system is Guugu Yimithirr, spoken in the Hope Vale region, northern Queensland, Australia. This language uses cardinal directions (north, south, east, west) for every single context related to position or direction. Thus, the speakers of this language do not talk about their ‘left’ or ‘right leg’, but rather their ‘west leg’ (meaning the right leg, if the speaker faces south or the left leg if the speaker faces north), ‘north leg’, etc.\n\nA special category of absolute referential systems, which is also the one most commonly appearing in linguistics problems, is that in which the reference system is based on the topography of the area. Usually, these words refer to directions such as ‘upstream’, ‘downstream’, ‘uphill’, ‘downhill’, ‘towards the forest’, ‘towards the shore’, etc.\n\nAs previously mentioned, with this type of problem, we need to pay attention to the topography of the area and consider the landforms around and how they can be used to indicate directions. In this case, since the language is spoken on an island, it is very plausible that the sea is one of the points of reference. Thus, each point can be closer to the sea or more towards the centre of the island (in this case the middle of the island is actually a volcano so equivalent directions can also be uphill and downhill). Since in the introduction we are told that the island has an active volcano, we expect it to be relevant in solving the problem. Considering that a volcano can be assimilated to a cone, we can consider the second direction to be circular (clockwise or anticlockwise).","source_file":"chapters/10-OtherProblems.tex","source_line_start":7,"license":"CC-BY-4.0","retrieval_text":"Other types of problems\nProblems based on orientation systems\norientation, kinship, and other structural problems\nThe purpose of orientation system problems is to identify the way in which a specific language expresses directions (relative positions of objects, such as in ‘in front’, ‘behind’, ‘to the left’, ‘to the right’, etc. or directions towards something: ‘go ahead’, ‘turn left’, ‘turn right’, etc.). These problems are typically easy to recognise since they usually contain the image of a map or a similar diagram or picture.\n\nTypologically speaking, orientation systems can be classified into two categories: absolute or relative referential systems. In absolute referential systems, the directions are relative to one or more fixed points (e.g., cardinal directions or geographical locations). Probably the best-known language which uses an absolute orientation system is Guugu Yimithirr, spoken in the Hope Vale region, northern Queensland, Australia. This language uses cardinal directions (north, south, east, west) for every single context related to position or direction. Thus, the speakers of this language do not talk about their ‘left’ or ‘right leg’, but rather their ‘west leg’ (meaning the right leg, if the speaker faces south or the left leg if the speaker faces north), ‘north leg’, etc.\n\nA special category of absolute referential systems, which is also the one most commonly appearing in linguistics problems, is that in which the reference system is based on the topography of the area. Usually, these words refer to directions such as ‘upstream’, ‘downstream’, ‘uphill’, ‘downhill’, ‘towards the forest’, ‘towards the shore’, etc.\n\nAs previously mentioned, with this type of problem, we need to pay attention to the topography of the area and consider the landforms around and how they can be used to indicate directions. In this case, since the language is spoken on an island, it is very plausible that the sea is one of the points of reference. Thus, each point can be closer to the sea or more towards the centre of the island (in this case the middle of the island is actually a volcano so equivalent directions can also be uphill and downhill). Since in the introduction we are told that the island has an active volcano, we expect it to be relevant in solving the problem. Considering that a volcano can be assimilated to a cone, we can consider the second direction to be circular (clockwise or anticlockwise)."}
{"id":"book_method_c10_s02_p01","kind":"method","chapter":10,"chapter_title":"Other types of problems","section":2,"section_title":"Kinship problems","topic":"orientation, kinship, and other structural problems","text":"The core purpose of this type of problem is that different languages use different types of terms to refer to different family relations. This type of problem is very easy to recognise since it will refer to a family tree. In the corpus, some sentences are given in which the relations of some family members with other members of the family are described in the target language.\n\nThere are six main types of kinship systems, but before discussing them, it is important to get acquainted with the ways of representing a kinship diagram (a family tree). Figure fig:short-kinship shows a basic kinship diagram. Triangles represent men and circles represent women. A double line (an equals sign, =) denotes marriage and simple lines represent blood relations. Figure fig:short-kinship tells us that 1 and 3 are two women (they are represented by circles) who are sisters (marked by a simple line), and each of them is married (1 married to 2 and 3 to 4). The family formed by 1 and 2 has two children (vertical line) – a girl (5) and a boy (6), just like the family formed by 3 and 4. Thus, based on this diagram, we can characterise each person with respect to the other, e.g., 6 is the son of 1 and the nephew of 4, 2 is the husband of 1 and the brother-in-law of 3, etc. Generally, linguistics problems will also include a legend which explains the meaning of each symbol, but the notations mentioned above are the standard ones. Moreover, from case to case, some problems might also include the age of the person, since some languages differentiate certain kinship terms based on age (e.g., in Chinese there are different words for ‘younger sister’ – 妹妹 mèimei, ‘older sister’ – 姐姐 jiějie, ‘younger brother’ – 弟弟 dìdi, and ‘older brother’ – 哥哥 gēge).\n\n[VISUAL OMITTED: figures/kinship_short.pdf]\nCaption: A typical kinship diagram.\n\nIn 1949 the anthropologist G.P. Murdock identified six basic patterns of kinship terminology systems, which are now generally accepted. Of course, certain languages can display systems different from them, but the six types below are the most common ones.\n\nIn order to represent these systems, a diagram like the one in Figure fig:template-kinship is used. A shaded triangle or circle, if included, represents the person from whose point of view the tree is presented, called the Ego in genealogy.\n\n[VISUAL OMITTED: figures/kinship_template.pdf]\nCaption: Template kinship diagram for describing the different types of kinship systems.\n\nThe simplest kinship system is called Hawaiian, which has only three basic kinship terms: ‘mother’, ‘father’ and ‘sibling’. Thus, persons A, D and E are all called ‘father’, persons B, C, F are all called ‘mother’ and all siblings and cousins are called the same (‘sibling’). In this case (the Hawaiian system), we write A = D = E, B = C = F and G = H = I = J = K = L = M = N = O = P. This representation does not take into account the variations based on age so, solely based on this description, we cannot know that in the Hawaiian language (which displays the Hawaiian kinship term) there is a difference between younger and older sibling. Generally, in linguistics problems, if the age does not play a relevant role, it will not be included in the diagram (i.e., if a diagram includes the age of the persons, age will certainly play a role).","source_file":"chapters/10-OtherProblems.tex","source_line_start":83,"license":"CC-BY-4.0","retrieval_text":"Other types of problems\nKinship problems\norientation, kinship, and other structural problems\nThe core purpose of this type of problem is that different languages use different types of terms to refer to different family relations. This type of problem is very easy to recognise since it will refer to a family tree. In the corpus, some sentences are given in which the relations of some family members with other members of the family are described in the target language.\n\nThere are six main types of kinship systems, but before discussing them, it is important to get acquainted with the ways of representing a kinship diagram (a family tree). Figure fig:short-kinship shows a basic kinship diagram. Triangles represent men and circles represent women. A double line (an equals sign, =) denotes marriage and simple lines represent blood relations. Figure fig:short-kinship tells us that 1 and 3 are two women (they are represented by circles) who are sisters (marked by a simple line), and each of them is married (1 married to 2 and 3 to 4). The family formed by 1 and 2 has two children (vertical line) – a girl (5) and a boy (6), just like the family formed by 3 and 4. Thus, based on this diagram, we can characterise each person with respect to the other, e.g., 6 is the son of 1 and the nephew of 4, 2 is the husband of 1 and the brother-in-law of 3, etc. Generally, linguistics problems will also include a legend which explains the meaning of each symbol, but the notations mentioned above are the standard ones. Moreover, from case to case, some problems might also include the age of the person, since some languages differentiate certain kinship terms based on age (e.g., in Chinese there are different words for ‘younger sister’ – 妹妹 mèimei, ‘older sister’ – 姐姐 jiějie, ‘younger brother’ – 弟弟 dìdi, and ‘older brother’ – 哥哥 gēge).\n\n[VISUAL OMITTED: figures/kinship_short.pdf]\nCaption: A typical kinship diagram.\n\nIn 1949 the anthropologist G.P. Murdock identified six basic patterns of kinship terminology systems, which are now generally accepted. Of course, certain languages can display systems different from them, but the six types below are the most common ones.\n\nIn order to represent these systems, a diagram like the one in Figure fig:template-kinship is used. A shaded triangle or circle, if included, represents the person from whose point of view the tree is presented, called the Ego in genealogy.\n\n[VISUAL OMITTED: figures/kinship_template.pdf]\nCaption: Template kinship diagram for describing the different types of kinship systems.\n\nThe simplest kinship system is called Hawaiian, which has only three basic kinship terms: ‘mother’, ‘father’ and ‘sibling’. Thus, persons A, D and E are all called ‘father’, persons B, C, F are all called ‘mother’ and all siblings and cousins are called the same (‘sibling’). In this case (the Hawaiian system), we write A = D = E, B = C = F and G = H = I = J = K = L = M = N = O = P. This representation does not take into account the variations based on age so, solely based on this description, we cannot know that in the Hawaiian language (which displays the Hawaiian kinship term) there is a difference between younger and older sibling. Generally, in linguistics problems, if the age does not play a relevant role, it will not be included in the diagram (i.e., if a diagram includes the age of the persons, age will certainly play a role)."}
{"id":"book_method_c10_s02_p02","kind":"method","chapter":10,"chapter_title":"Other types of problems","section":2,"section_title":"Kinship problems","topic":"orientation, kinship, and other structural problems","text":"The simplest kinship system is called Hawaiian, which has only three basic kinship terms: ‘mother’, ‘father’ and ‘sibling’. Thus, persons A, D and E are all called ‘father’, persons B, C, F are all called ‘mother’ and all siblings and cousins are called the same (‘sibling’). In this case (the Hawaiian system), we write A = D = E, B = C = F and G = H = I = J = K = L = M = N = O = P. This representation does not take into account the variations based on age so, solely based on this description, we cannot know that in the Hawaiian language (which displays the Hawaiian kinship term) there is a difference between younger and older sibling. Generally, in linguistics problems, if the age does not play a relevant role, it will not be included in the diagram (i.e., if a diagram includes the age of the persons, age will certainly play a role).\n\nThe next system is called EskimoThe term dates from 1949 and is still used even though the word Eskimo is now disdained as being derogatory. kinship (or Inuit) one, which is also the system used in English. In this system we differentiate C (‘mother’), D (‘father’), B = F (‘aunt’), A = E (‘uncle’), K = L (‘siblings’) and G = H = I = J = M = N = O = P (‘cousins’). Although the six main kinship systems we describe here represent general patterns, these patterns can be slightly modified from one language to another. A case worth mentioning is that of Romanian which, although it is considered to use an Eskimo kinship term, it has two different words for ‘cousin’ based on the gender, and differentiates between G = I = M = O (verișoară, ‘female cousin’) and H = J = N = P (verișor, ‘male cousin’). This is true of many European languages, and in some (e.g. German) the difference Kusine–Vetter is not just a gender suffix.\n\nThe next system we talk about is called Sudanese kinship, one example of which is Turkish, which we use here to illustrate. In this type of system there is a separate term for each of the persons A-F (dayı = ‘mother's brother’, amca = ‘father's brother’, teyze = ‘mother's sister’, hala = ‘father's sister’, anne = ‘mother’, baba = ‘father’) and a different term for each pair of cousins: I and J are called ‘maternal parallel cousins’ – maternal refers to the fact that they are on the mother's side. The term parallel is used because the blood relation is from the same-sex persons (i.e., same-sex siblings) - mother and mother's sister (two women). In the same way, O and P are called ‘paternal parallel cousins’, from the father's side and, more exactly, from the father's brother (parallel since it is the father's brother (same sex as the father), not sister). The other two categories are called ‘maternal cross cousins’ and ‘paternal cross cousins’, where cross refers to the fact that the blood relation is of opposite-sex persons (mother's brother and father's sister).\n\nThis distinction between parallel and cross cousins is rather common and so relevant that in the next kinship system, called Iroquois kinship, B = C (‘mother’) and D = E (‘father’). Here, the same-sex siblings of the parents (i.e., mother's sister and father's brother) are also considered to be `parents'. On the other hand, the opposite-sex siblings of the parents are those called ‘uncle’ (A) and ‘aunt’ (F). For this reason, G = H = O = P (‘cousins’) – since they are the children of the aunt and uncle –, but I = J = K = L = M = N (‘siblings’) – since they are the children of the mother and father. The next two systems are derived from this system.","source_file":"chapters/10-OtherProblems.tex","source_line_start":83,"license":"CC-BY-4.0","retrieval_text":"Other types of problems\nKinship problems\norientation, kinship, and other structural problems\nThe simplest kinship system is called Hawaiian, which has only three basic kinship terms: ‘mother’, ‘father’ and ‘sibling’. Thus, persons A, D and E are all called ‘father’, persons B, C, F are all called ‘mother’ and all siblings and cousins are called the same (‘sibling’). In this case (the Hawaiian system), we write A = D = E, B = C = F and G = H = I = J = K = L = M = N = O = P. This representation does not take into account the variations based on age so, solely based on this description, we cannot know that in the Hawaiian language (which displays the Hawaiian kinship term) there is a difference between younger and older sibling. Generally, in linguistics problems, if the age does not play a relevant role, it will not be included in the diagram (i.e., if a diagram includes the age of the persons, age will certainly play a role).\n\nThe next system is called EskimoThe term dates from 1949 and is still used even though the word Eskimo is now disdained as being derogatory. kinship (or Inuit) one, which is also the system used in English. In this system we differentiate C (‘mother’), D (‘father’), B = F (‘aunt’), A = E (‘uncle’), K = L (‘siblings’) and G = H = I = J = M = N = O = P (‘cousins’). Although the six main kinship systems we describe here represent general patterns, these patterns can be slightly modified from one language to another. A case worth mentioning is that of Romanian which, although it is considered to use an Eskimo kinship term, it has two different words for ‘cousin’ based on the gender, and differentiates between G = I = M = O (verișoară, ‘female cousin’) and H = J = N = P (verișor, ‘male cousin’). This is true of many European languages, and in some (e.g. German) the difference Kusine–Vetter is not just a gender suffix.\n\nThe next system we talk about is called Sudanese kinship, one example of which is Turkish, which we use here to illustrate. In this type of system there is a separate term for each of the persons A-F (dayı = ‘mother's brother’, amca = ‘father's brother’, teyze = ‘mother's sister’, hala = ‘father's sister’, anne = ‘mother’, baba = ‘father’) and a different term for each pair of cousins: I and J are called ‘maternal parallel cousins’ – maternal refers to the fact that they are on the mother's side. The term parallel is used because the blood relation is from the same-sex persons (i.e., same-sex siblings) - mother and mother's sister (two women). In the same way, O and P are called ‘paternal parallel cousins’, from the father's side and, more exactly, from the father's brother (parallel since it is the father's brother (same sex as the father), not sister). The other two categories are called ‘maternal cross cousins’ and ‘paternal cross cousins’, where cross refers to the fact that the blood relation is of opposite-sex persons (mother's brother and father's sister).\n\nThis distinction between parallel and cross cousins is rather common and so relevant that in the next kinship system, called Iroquois kinship, B = C (‘mother’) and D = E (‘father’). Here, the same-sex siblings of the parents (i.e., mother's sister and father's brother) are also considered to be `parents'. On the other hand, the opposite-sex siblings of the parents are those called ‘uncle’ (A) and ‘aunt’ (F). For this reason, G = H = O = P (‘cousins’) – since they are the children of the aunt and uncle –, but I = J = K = L = M = N (‘siblings’) – since they are the children of the mother and father. The next two systems are derived from this system."}
{"id":"book_method_c10_s02_p03","kind":"method","chapter":10,"chapter_title":"Other types of problems","section":2,"section_title":"Kinship problems","topic":"orientation, kinship, and other structural problems","text":"This distinction between parallel and cross cousins is rather common and so relevant that in the next kinship system, called Iroquois kinship, B = C (‘mother’) and D = E (‘father’). Here, the same-sex siblings of the parents (i.e., mother's sister and father's brother) are also considered to be `parents'. On the other hand, the opposite-sex siblings of the parents are those called ‘uncle’ (A) and ‘aunt’ (F). For this reason, G = H = O = P (‘cousins’) – since they are the children of the aunt and uncle –, but I = J = K = L = M = N (‘siblings’) – since they are the children of the mother and father. The next two systems are derived from this system.\n\nThe Crow kinship system starts from the Iroquois system, with the only change occurring for the persons O and P (the children of the father's sister). They are called ‘aunt’ (if it is a girl, so O) or ‘father’ (P). Thus, in this system, D = E = P (‘father’) and F = O (‘aunt’). The rest of the persons follow the Iroquois system (A = ‘uncle’, B = C = ‘mother’, G = H = ‘cousins’, I = J = K = L = M = N = ‘siblings’).\n\nThe last system, called Omaha kinship, is the opposite of Crow kinship. In this system, a special role is reserved for the children of the mother's brother (instead of father's sister, as in the Crow system). Like the Crow system, the same-sex child is also called ‘uncle’ or ‘aunt’, depending on their (and Ego's) sex (same sex as Ego), while the opposite-sex child is called ‘mother’ or ‘father’. Therefore, A = H = ‘uncle’, B = C = G = ‘mother’, D = E = ‘father’, F = ‘aunt’, I = J = K = L = M = N = ‘sibling’ and O = P = ‘cousins’.\n\nAlthough the systems are each named after a language, there are other languages which follow each system: for example English has `Eskimo kinship', Bulgarian has `Sudanese kinship'.\n\nA comparative representation of these six types of system is represented below:\n\nCC = ‘cross cousin’, PC = ‘parallel cousin’.","source_file":"chapters/10-OtherProblems.tex","source_line_start":83,"license":"CC-BY-4.0","retrieval_text":"Other types of problems\nKinship problems\norientation, kinship, and other structural problems\nThis distinction between parallel and cross cousins is rather common and so relevant that in the next kinship system, called Iroquois kinship, B = C (‘mother’) and D = E (‘father’). Here, the same-sex siblings of the parents (i.e., mother's sister and father's brother) are also considered to be `parents'. On the other hand, the opposite-sex siblings of the parents are those called ‘uncle’ (A) and ‘aunt’ (F). For this reason, G = H = O = P (‘cousins’) – since they are the children of the aunt and uncle –, but I = J = K = L = M = N (‘siblings’) – since they are the children of the mother and father. The next two systems are derived from this system.\n\nThe Crow kinship system starts from the Iroquois system, with the only change occurring for the persons O and P (the children of the father's sister). They are called ‘aunt’ (if it is a girl, so O) or ‘father’ (P). Thus, in this system, D = E = P (‘father’) and F = O (‘aunt’). The rest of the persons follow the Iroquois system (A = ‘uncle’, B = C = ‘mother’, G = H = ‘cousins’, I = J = K = L = M = N = ‘siblings’).\n\nThe last system, called Omaha kinship, is the opposite of Crow kinship. In this system, a special role is reserved for the children of the mother's brother (instead of father's sister, as in the Crow system). Like the Crow system, the same-sex child is also called ‘uncle’ or ‘aunt’, depending on their (and Ego's) sex (same sex as Ego), while the opposite-sex child is called ‘mother’ or ‘father’. Therefore, A = H = ‘uncle’, B = C = G = ‘mother’, D = E = ‘father’, F = ‘aunt’, I = J = K = L = M = N = ‘sibling’ and O = P = ‘cousins’.\n\nAlthough the systems are each named after a language, there are other languages which follow each system: for example English has `Eskimo kinship', Bulgarian has `Sudanese kinship'.\n\nA comparative representation of these six types of system is represented below:\n\nCC = ‘cross cousin’, PC = ‘parallel cousin’."}