Matthew Dryer

The data is organized by language, the data for each language separated by a blank line, the languages being organized alphabetically. Each line (though occasionally spilling over onto another line) is associated with one parameter. Parameters beginning with a capital letter are nonlinguistic parameters: L. for language name (N.is alternative, for languages without sufficient data at least as yet), S. for source, G for rough information on genetic classification (according to Voegelin and Voegelin (1977), Ruhlen (1987), or Ethnologue), GE. for genus name, A. for continental area (given only for the FIRST language in the database in that area), F. for family (given only for the first lg in that family), P. for latitude and longtitude (P stands for position), and PL. for place, i.e. the name of a country or part of a country, X. for subsample that this language is included in (1=primary sample (implicit, equivalent to ‘L.’ in first field), 4=first 100 languages for atlas, 5=second 100 languages for atlast, 6=200 languages for atlas, 7=languages for Dryer maps in atlas, 8=additional languages for atlas from sparser areas), E. for three-letter code used by Ethnologue, B. for three-letter code used on maps if different from first three letters of language name, D. for date done. The first line is always the language name in capital letters. This line may contain information indicating comparison with the data in Hawkins (1983) (John Hawkins, Word Order Universals , Academic Press). "Hn" indicates that the language is not listed by Hawkins (at least not with the same name). "Hc" indicates that the language is one listed listed by Hawkins. Where Hawkins lists characteristics not listed in the data base, the additional characteristics are mentioned in this place. Where there are contradictions between Hawkins and this data base, mention is made under the characteristic. The source entry contains a list of the bibliographic sources. In some (though not all) cases in which more than one source was used, one source is treated as the primary source, other sources as secondary sources. In such cases, the secondary sources are listed in parentheses, precede by "Also:". Page numbers listed with characteristics without mention of source are from the primary source.

The remainder of each entry for a language is a listing of the available typological characteristics, each line starting with a number indicating the word order parameter. Where the first space on a line is blank, the number is assumed to be the same as the previous line. Some parameters are subdivided into a number of subcases, distinguished by letter. The absence of a letter is equivalent to the letter "a".

Each line begins with the number of the parameter followed (in some cases) by a letter indicating the parameter, followed by an abbreviation of the value of the parameter for that language. These are discussed below. This value is followed by a semi-colon, which is followed by an indication of the specific source, usually by page number with the nature of the evidence in parentheses. When there is a single source used or a primary source specified, page numbers are assumed to be in that source. In other cases, the page numbers are preceded by the name of the author of the source, followed by a colon. The abbreviations used for the nature of the evidence are as follows: st(s) = statement(s), eg(s) = example(s), imp = implied, tc text count, elic = elicitation. Occasionally, additional information of interest is added at the end of the line, including "(contra Hawkins)" indicating Hawkins (1983) gives a different value for this parameter.

A number of comments apply to different parameters. Many characteristics include (B) for "bound" (i.e. an affix), (NB) for "not bound" (either a separate word or at any rate not bound to the item in question), or (CL) for clitic. This last value is appropriate when the item is like an affix in combining phonologically with some word to form a single word, but where its position is determined syntactically in that it combines syntactically with a phrase rather than to the word. A morpheme that combines with whatever is the last word in a phrase regardless of the word class of the word it combine with, like possessive "'s" in English, counts as a clitic. Where more than one of these two possibilities exist within the same language, both values are listed, separated by "/", e.g. (B/NB).

In general all-lower case entries indicate that there is evidence that a given value is possible (most commonly by an isolated example), but possibly not basic. Thus for 2a (order of noun and adjective), the following values are possible:

The difference between AN/NA and an/na is somewhat vague. AN/NA would be used where there is lots of evidence that both orders occur, and that it seems unlikely that any convincing argument might be given that one is more basic. an/na would be used where there is relatively little evidence, where additional evidence might place the language in AN/na, an/NA or AN/NA. These distinctions are relevant to other parameters as well, but the lower case possibilities will not be mentioned below. Note that in previous stages of the project, the only values used were AN and NA, and for many of the languages, these are the only kinds of values used.

The parameters with their possible values are as follows. Values that not technically allowed or which are not recognized at this time by the computer program, or which I now try to avoid using, are enclosed in parentheses (e.g. SNegVNegO for 16b).

0.(a.) (NB: 0.=0.a.) The basic order of nominal subject, nominal object, and verb.

The value "Rigid" can only be used in combination with other parameters: SOV/Rigid means that the language is rigidly SOV. By itself, Flex is equivalent to SOV/OSV/SVO/OVS/VSO/VOS. Examples of possible values:

0.b. The basic order of subject and verb in the absence of conclusive evidence of the basic position for objects.

0.c. The basic order of object and verb in the absence of conclusive evidence of the basic position of subjects.

0.d. The basic order of subject and object in the absence of conclusive evidence of the basic position for the verb.

0.e. The basic order of intransitive subject with respect to verb or transitive subject with respect to verb, where there is a difference between the two kinds of subjects (if there position is the same, that information goes under (b)). For example, a language which is basically SVO for transitive clauses but which allows both orders equally for intransitive subjects would be listed as SVO under (a) and as AV/SiV/VSi here under (e).

1.(a.) The basic order of genitive (G) and noun (N).

1.b. Type of structure used in genitive construction.

2.(a.) The basic order of attributive (modifying) adjective (A) and noun (N)

2.b. Information on the word class of adjectives; are they verbs?

2.c. Information on the word class of adjectives; are they nouns?

2.d. Information on the word class of adjectives; do they

take a copula?

3.(a.) The basic order of adposition and NP (i.e. whether the language is prepositional or postpositional)

3.b. The categorial status of adpositions.

4.(a.) The basic order of demonstrative (Dem) and noun (N).

4.b. The basic order of article (Art) and noun (N). This parameter has been superceded for most languages by (4c) and (4d). But it is also used for articles that do not code definiteness or indefiniteness but which they serve as markers of NP's and are often described as articles in grammars. If there are no articles, the entry NoArt here under 4b suffices, implying NoDef under (4c) and NoIndef under 4d).

4.c. The basic order of definite article (Def) and noun (N). Articles coding specificity rather than definiteness are treated as definite articles.

4.d. The basic order of indefinite article (Indef) and noun (N). Morphemes indicating nonspecific are treated here as indefinite articles.

4.e. The basic order of classifier (Clsfr) and noun (N).

4.f. Is the definite article the same as the demonstrative?

4.g. Is the indefinite article the numeral "one"?

4.h. Can articles co-occur with demonstratives?

4.i. Forms for demonstratives (demonstrative modifiers)

4.j. Are demonstrative modifiers the same in form as demonstrative pronouns?

5.(a.) The basic order of numeral (Num) and noun (N)

5.b. The basic order of numeral and classifier.

5.c. The basic order of ordinal numeral and noun.

5.d. Forms for numerals ‘one’ to ‘ten’.

6.(a.) The basic order of relative clause (Rel) and head noun (N).

b. Structure of relative clause

6.c. Position of relative marking relative to relative clause. These are mostly merged into 6.b.

7.(a.) The basic order of noun stem and case affix (i.e. whether the language uses case suffixes or case prefixes).

7.b. The nature of case marking of subjects. (Note that "case marking" is used here to include either case affixes or adpositions)

7.c. The nature of case marking of (direct) objects.

7.d. The nature of case marking of indirect objects (IO).

7.e. Not currently used.

7.f. Not currently used.

7.g. Primary object case marking (IOs and DOs in monotransitive clauses). No longer used.

7.h. No longer used.

8.(a.) The basic order of manner adverb (Adv) and verb (V)

8.b. Position of adverb relative to subject, object, verb.

9.(a.) The basic position of pronominal subject affixes on the verb

9.b. Further information on subject affixes with respect to whether they code person, number, or gender.

This field was formerly used for status of pronominal subject affixes with respect to role at clause level. These are being replaced by 9.f., 9.g., and 9.h. The values that were used included

9.c. Position of independent subject pronouns relative to the verb.

9.d. Status of independent subject pronouns with respect to role at clause level

9.e. Forms of independent pronouns

9.f. Are independent pronouns for subject in subject position obligatory or usually used?

9.g. Are there pronominal subject morphemes other than pronouns occurring in subject position?

9.h. Do pronominal morphemes which are distinct from pronouns in subject position (i.e. those fitting categories other than OnlySuPro in 9g) normally co-occur with full noun phrase subjects?

10.(a.) The basic position of pronominal object affixes on the verb

10.b. Further information on object affixes with respect to whether they code person, number, or gender.

10.c. etc. Analogous to 9.c. etc. but for objects.

11.(a.) The basic position of "auxiliaries" (Aux) (i.e. morphemes not bound to verb indicating tense or aspect) relative to the lexical verb (V)

11.b. Position of Aux relative to subject, object, verb.

11.c. Categorial status of auxiliary.

12. The basic position of tense/aspect affixes relative to the verb stem

13.(a.) The basic position of modality morphemes (Mod) (indicating ability, permission, possibility, or obligation) relative to the lexical verb (or verb stem) (V). If the only data is on morphemes indicating ability, it is sufficient to list this information only under (13d)

13. b. Categorial status of modals that are NB. Where there are different modals with different statuses, the value relative to a modal indicating ability should be listed here first.

13. c. Page numbers where there is information on other "modal" morphemes, including ones indicating doubt, certainty, hearsay,etc.

13.d. The basic position of morphemes indicating ability relative to the lexical verb (or verb stem).

13.e. Categorial status of words meaning ‘able’.

14. The basic position of morphemes indicating desire of subject (Want) relative to the lexical verb or verb stem (V) denoting the event wanted. (Do NOT include optative morphemes that indicate desire of speaker rather the desire of subject.)

15. The basic position of causative morphemes (Caus) relative to lexical verb or verb stem (V)

16.(a.) The basic position of negative morphemes (Neg) relative to main/lexical verb or verb stem (V)

16.b. Position of negative relative to subject, object, verb.

16.c. Categorial status of negative that is NB.

16.d. Order of negative and tense-aspect auxiliary.

16.e. Order of nonverbal predicate and negative copula

17.(a.) The basic order of adjective (Adj) and degree word (intensifier, qualifying word) (Deg) (like very in English) (including bound morphemes expressing same meaning).

17.b. The basic order of adverb (Adv) and degree word (Deg).

18.(a.) The basic order of adjective (Adj), marker (M), and standard (St) in comparatives

18.b. The basic order of marker (M) and standard (St)

18.c. The basic order of adjective (Adj) and standard (St)

19. The basic position of question particles or affixes (used to indicate a yes/no question) (Q) relative to the main verb or verb stem (V)

20. Whether interrogative words in content questions obligatorily occur in sentence-initial position or not

21. a. The basic position of adverbial subordinators (morphemes indicating specific semantic relationships of adverbial clauses to the main clause, with meanings such as ‘when’, ‘because’, ‘after’, ‘although’, ‘if’) relative to the rest of the subordinate clause

21.b. The basic order of adverbial subordinate clause and main clause. This should not be based on the order of conditional clause and main clause, since these are often initial when other subordinate clauses are not.

21.c. The basic order of purpose complement with respect to main verb. (E.g. in English ‘He went home to eat dinner’, ‘went’ is main verb and ‘to eat (dinner)’ is purpose complement.

22.(a.) The basic order of direct objects (DO) relative to indirect objects (IO)

22.b. Marking of indirect object, if that information is not derivable from 7d.

23. The basic order of complementizer and clause. Note that this number was originally used for information that should now be included under category 26.

24.(a.) The basic order of copulative morphemes (Cop) (equivalent to English "be" or affixes with a similar function) and predicates (Pred). Note that at earlier stages of the project data was included here for other copulas, information that should now be listed under (24d).

24.b. The categorial status of copulas which are NB.

24.c. Do adjective predicates occur with a copula?

24.d. Do noun predicates occur with a copula?

24.e. Do locative predicates occur with a copula?

24.f. Basic order of morpheme meaning "become" relative to predicate. (This parameter was originally used for other copulas in general with values OtherCopPred and PredOtherCop.)

25.(a.) The basic order of adpositional phrases (PP) relative to the verb (V)

25.b. The basic order of oblique nominals (including adpositional phrases) (X) relative to the verb (V). Values should be listed here only if there is no value under (25a).

25. c. The basic order of oblique nominals (including adpositional phrases) (X) relative to the subject (S), object (O), and verb (V)

26. The basic order among modifiers of the noun (N), including adjectives (A), numerals (Num), demonstratives (Dem), articles (Art), and genitives (G) Use commas to separate modifiers that occur in same slot (either because they are mutually exclusive or because they occur in either order.)

27. The basic order among verbal affixes, primarily those indicating tense/aspect (T,TA,A), subject (Su), and object (Ob)

28.(a.) Basic order of interrogative words meaning "which" and noun. Note that this number was originally used for another purpose, data that should be ignored.

28.b. Basic order of other modifying interrogative words relative to noun..Mention of other interrogative words that modify nouns, such as interrogative possessives (IntPoss, like English ‘whose’), interrogative numerals (IntNum, like English ‘how many’), or true interrogative adjectives IntAdj, like English ‘what sort of’). (Note that this number was originally used for another purpose, data that should be ignored)

29. Basic order of universal quantifier ("all", "every", "each") and noun. Note that this number was originally used for another purpose, data that should be ignored.

30.(a.) The basic order of plural words or affixes (Pl) relative to the noun (N). (30b, 30c, and 30d apply only to plural words.)

30.b. Order of plural word with respect to other modifiers that occur on same side of noun. (This information should also be incorporated into 26.)

30.c. Category status of plural word.

30.d. Other information on plural words, on whether numerals can co-occur with plural words, on what else is coded by plural word (e.g.gender or definiteness), and on whether there are also singular words or dual words. (Only applies to plural words, not plural affixes.)

31. a. The basic position of pronominal affixes on possessed nouns indicating person and/or number of possessor. (The previous values used here were PossPref and PossSuff.)

31.b. The basic position of possessive words, pronominal words indicating person and/or number of possessor modifying nouns.

32. Do adjectives agree with nouns they modify?

33. Construction used for ‘it is raining’

34. Construction for saying ‘X has Y’

The set of parameters for which data has been collected has gradually grown during the project. As a result, the earlier that data was collected, the smaller the set of parameters for which data was collected (unless I have gone back and reexamined the language, which I have often done, especially for certain languages). The following is an approximate schedule in which data was collected. In general if there is data for a later parameter, that means the language has been examined up to that point. In many cases, parameter values for a given parameter have increased over time (especially in adding ‘No’ values, like ‘NoCaseAff’, ‘NoArt’, etc.) Some parameters have not had data collected for them, even though they were instituted earlier.

15 August 2000