EDITORS: Geoffrey Sampson, David Gil, and Peter Trudgill TITLE: Language Complexity as an Evolving Variable SERIES: Studies in the Evolution of Language 13 PUBLISHER: Oxford University Press YEAR: 2009
Grover Hudson, Department of Linguistics and Germanic, Slavic, Asian and African Languages, Michigan State University
''This book is the outcome of a workshop held at the Max Planck Institute for Evolutionary Anthropology, Leipzig, in April 2007 ... convened in response to the fairly sudden emergence, in many diverse quarters internationally, of scepticism about a longstanding linguistic axiom - that all languages and language varieties are similar in complexity, and have been so at all times in the past'' (vii). Important work 'in many diverse quarters' to which authors here respond includes Givón 1979, Hawkins and Gell-Mann, eds. 1992, Trudgill 1998, Deutscher 2000, McWhorter 2001, Kusters 2003, Dahl 2004, Everett 2005, and Miestamo et al., eds. 2008.
There are nineteen chapters, as follows.
1. ''A linguistic axiom challenged'' by Geoffrey Sampson is an overview of the book-title claim. The opposite claim is more widely held, according to Sampson, e.g. Hockett (1958: 180): ''impressionistically, it would seem that the total grammatical complexity of any language, counting both morphology and syntax, is about the same as that of any other. This is not surprising, since all languages have about equally complex jobs to do, and what is not done morphologically has to be done syntactically.'' As a counterexample to the claim of equal complexity, Sampson raises the extraordinarily elaborate morphology of the Caucasian language Archi, described by Kibrik (1998). Sampson quotes Jackendoff (1993: 32) that ''the earliest written documents already display the full expressive variety and grammatical complexity of modern languages.'' He reviews recent work which begins to challenge the traditional idea, including McWhorter (2001) that creoles are characterized by lesser complexity than other languages, and Everett (2005) that Pirahã is a language without (among other things) recursion.
2. ''How much grammar does it take to sail a boat?'' by David Gil continues his argument that Rau Indonesian represents a ''relative IMA (isolating-monocategorical-associational) language,'' having isolating morphology, few if any part-of-speech categories, and little more than pragmatic interpretation of word associations (22). The chapter title relates to the hypothesis that early-human colonizers of Austronesia had no more than an IMA to build and sail the boats they needed to reach their islands. According to Gil, ''pure IMA Language is endowed with substantial expressive power'' (29), and the complexity of other languages is largely superfluous anyway, as ''the fact remains that language is hugely disfunctional. Just think of all the things that it would be wonderful to be able to say but for which no language comes remotely near to providing the necessary expressive tools'' (32).
3. ''On the evolution of complexity: sometimes less is more in East and mainland Southeast Asia'' by Walter Bisang argues that absence of complexity in the usual sense may result in 'hidden complexity': increased need for inference. ''Hidden complexity reflects economy: the structure of the language does not force the speaker to use a certain grammatical category if it can be inferred from context'' (34). Gil's IMA languages ''are not necessarily simple''; ''their overt structural simplicity may mask hidden complexity'' (35). Most of the chapter is exemplification of southeast Asian language ''markers [that] have a wide range of meaning,'' and ''utterances that may represent a number of different constructions'' (38).
4. ''Testing the assumption of complexity invariance: the case of Elfdalian and Swedish'' by Östen Dahl compares the two ''North Germanic languages ... to identify the points at which the languages differ in complexity and see whether these differences are compensated for elsewhere in the grammar'' (50). Elfdalian is ''an endangered vernacular spoken by about 3000 persons ... not mutually intelligible'' with Swedish. Dahl distinguishes 'absolute' and 'agent-related' complexity. Absolute complexity of an object ''is related to the amount of information needed to recreate or specify it'' or ''the length of the shortest possible complete description of it'' (50); he is unconcerned with 'agent-related complexity', which is learners' ''difficulty in learning, producing, and understanding the language'' (51). According to his comparisons, ''it is in fact reasonable to assume that Elfdalian has a more complex grammar than Swedish'' (63).
5. ''Between simplification and complexification: non-standard varieties of English around the world'' by Benedikt Szmrecsanyi and Bernd Kortmann is an ''endeavour to analyse and interpret language-internal variation in English in terms of varying complexity and simplicity levels'' (64). They employ four comparisons (three of morphological complexity plus 'L2 acquisition difficulty') on four varieties of English: 'high-contact' vernacular Australian, 'low-contact' vernacular East Anglian, English-based pidgin/creole Tok Pisin, and L2 English of Hong Kong. They conclude that ''variety type is a powerful predictor of complexity variance in varieties of English around the world ... Low-contact, traditional L1 vernaculars are on almost every count ... more complex than high-contact L1 varieties'' (76).
6. ''Implicational hierarchies and grammatical complexity'' by Matti Miestamo ''examines the usability of Greenbergian implicational hierarchies ... in cross-linguistic research on language complexity ... [T]he higher a language on a given hierarchy'' (for example Trial>Dual>Plural>Singular) ''the more complex its grammar in that respect'' (80). ''The study is based on a sample of 50 languages'' distributed geographically and genetically (84) with regard to their fulfillment of the opposite hierarchies of 'agreement' (Sub/Abs < Obj/Erg < Obl) and 'case' (Sub/Abs > Obj/Erg > Obl) (88). (That is, for the agreement hierarchy, agreement with obliques implies agreement with objects or ergatives which implies agreement with subjects or absolutives.) Although ''correlations are to be expected between the agreement and case hierarchies,'' Miestamo found ''no significant correlation, positive or negative'' (95).
7. ''Sociolinguistic typology and complexification'' by Peter Trudgill reviews his work since 1983 arguing that languages differ in complexity according to (1) amount of other-language contact (less contact, more complexity - as argued in ch. 5); (2) density of social networks (greater density, more complexity); and (3) number of speakers (fewer speakers, more complexity - as argued in ch. 9). He considers the three factors independent if related, but believes that ''widespread adult-only language contact is ... a mainly modern phenomenon associated with the last 2,000 years,'' so ''the dominant standard modern languages in the world today are likely to be seriously atypical of how languages have been for nearly all of human history'' (109).
8. ''Linguistic complexity: a comprehensive definition and survey'' by Johanna Nichols is ''a first step'' toward a ''cross-linguistic survey of complexity levels'', ''giving a comprehensive definition of complexity that should make it possible to draw grammatical samples, and doing a large enough cross-linguistic survey of enough parts of grammar to indicate whether the assumption of equal complexity appears viable'' (111). Nichols' survey found ''no significant negative correlations between different components of grammar,'' but ''a significant positive correlation between complexity of syntax and complexity of [morphological] synthesis,'' [and] ''it is negative correlations that support the hypothesis of equal complexity'' (119). Nor did Nichols find evidence for ''a preferred complexity level or trade-off between complexity levels of different components of grammar'' (120).
9. ''Complexity in core argument marking and population size'' by Kaius Sinnemäki asks ''whether core argument marking shows complexity varying with speech community size,'' and ''test[s] this relationship statistically with a sample of fifty languages. Complexity is measured as violations of distinctiveness and economy, the two sides of the principle of one meaning-one form'' (126; that is, one meaning should have one form and one form, one meaning). Sinnemäki notes research by Hay and Bauer (2007), who ''tested the relationship between phoneme inventory size and speech community size in 216 languages, and arrived at a statistically very significant positive correlation'' (128). He hypothesizes that ''languages spoken by small speech communities are likely to violate the principle ... by either redundant or insufficient morphosyntactic marking of core arguments'' but ''languages spoken by large speech communities are likely to adhere to the principle'' (131), and finds ''rather strong support for both hypotheses'' (138), with the exception that many large 'Old World' languages of Eurasia, Africa, and SE Asia tolerate violations of economy ('redundancy': two(+) forms per meaning).
10. ''Oh nɔ́ɔ!: a bewilderingly multifunctional Saramaccan word teaches us how a creole language develops complexity'' by John McWhorter argues that the second word of the title, which comes from 'no more', is evolving as a focus (new information) morpheme. He believes ''the discussion will also serve as a corrective to a tendency in studies of grammatical complexity - in creoles and beyond - to conceive of complexity as essentially a shorthand for inflectional morphology'' (142).
11. ''Orality versus literacy as a dimension of complexity'' by Utz Maas argues that European vernacular languages evolved complexity in becoming literate. In the example of an extract from a bilingual Latin/German contract of 1251, ''the Latin text does not cause any problems of interpretation'' (167), but the equivalent German text ''will not have been comprehensible to a contemporary reader ... It took a long time to elaborate the vernacular languages so that they could articulate complex literate texts, and Latin served as the model'' (169). Developing ''language that allows one to articulate all registers ... makes the language more complex, compared to what it would be if it could only articulate the intimate register - but it makes linguistic competence less complex than in cases where different languages are needed for the different registers. Linguistic complexity is the price for linguistic integration of different registers'' (172).
12. ''Individual differences in processing complex grammatical structures'' by Ngoni Chipere asserts ''the ... widespread assumption'' (of questionable relevance to the argument of this book) ''that all native speakers possess a uniform underlying competence to process complex grammatical structures,'' and concludes ''this assumption is false'' (178). He mentions research by Miller and Isard (1964), who ''discovered that some individuals could process sentences with at most one level of self-embedding, whereas other individuals could process sentences with two levels of self-embedding'' (183). In Chipere's research 'high academic ability' students perform better than 'low academic ability' students in sentence recall and comprehension. ''Memory training caused increases in recall...but not in comprehension, [but] comprehension training...caused increases in both comprehension and recall,'' which he attributes to ''differences in grammatical competence and not ... differences in working memory'' (performance) (189).
13. ''Origin and maintenance of clausal embedding complexity'' by Fred Karlsson considers constraints on ''possibilities of repeatedly embedding subordinate clauses in various positions in their main clauses,'' and he ''inquires whether [the constraints] have been stable and fluctuating over time'' (192). ''The major expository genre'' before writing, oral narrative, ''has been shown to be aggregative and paratactic rather than subordinating'' (195). Karlsson refers to research of Givón (1979; see now Givón and Shibatani 2009) that ''more complex forms of (especially finite) clausal embedding arose [in languages] as part of large-scale grammaticalization, along with the advent of written language and the consequent gradual conventionalization of written registers'' (201). ''The upper limits of clausal embedding complexity,'' however, ''have remained the same since the advent of written language'' (202).
14. ''Layering of grammar: vestiges of protosyntax in present-day languages'' by Liliana Progovac has as ''the first goal ... to demonstrate that Root Small Clause ... syntax is measurably simpler than sentential TP syntax'' (212). A Root Small Clause is such as 'Him retire?', 'Me first!', and 'Problem solved'. She considers these to lack a ''Tense Phrase (TP), but also Move and structural (nominative) case, which are associated with TP'' (212). A second argument is that ''Root SCs (and exocentric compounds), which approximate this rudimentary grammar, can be seen as living fossils of a protosyntactic stage in language evolution''; this Progovac finds ''consistent with the possibility that there exist languages spoken today which are characterized by a version of Root SC syntax, and such languages may include'' Riau Indonesian and Pirahã (212).
15. ''An interview with Dan Everett'' by Geoffrey Sampson replaces Everett's talk at the workshop. The editors thought the interview ''might advance our understanding of the overall workshop topic better'' by ''challenging Everett on some of the points where linguists have felt sceptical'' (213). According to Everett, the Amazonian Pirahã ''have independently discovered the usefulness of living one day at a time'' and correspondingly have simple language lacking ''recursion, number, numerals, and counting'' (213). Everett supposes the language to represent an ''earlier stage of language evolution,'' in which ''parataxis and adjunction precede embedding'' (215). Sampson presses Everett on these points of controversy concerning both linguistic and cultural relativity. For the theoretical linguistic issues, see now Everett 2009.
16. ''Universals in language or cognition? Evidence from English language acquisition and from Pirahã'' by Eugénie Stapert ''compare[s] mental verb constructions in English and Pirahã ... [T]he syntax of the sentences at issue seems very different at first sight. However, I shall argue that the difference is perhaps not as profound as it seems... English mental verb constructions need not be analysed as complex structures in young children's speech ... which in many respects is similar to what we find in Pirahã'' (230-1). If Pirahã lacks recursion, how does it express the sentential complements of 'mental verbs' like 'think', 'notice', and 'doubt', which seem to be recursive clauses? Referring to research of Diessel and Tomasello (2001), Stapert observes that mental verbs are paralleled in meaning and use by adverbs like 'apparently' (='I think'), 'clearly' ('I notice'), 'maybe' ('I doubt'); child acquirers of English similarly understand and structure their early mental verbs (239).
17. '''Overall complexity': a wild goose chase'' by Guy Deutscher counters two basic arguments for the claim that all languages are equally complex ('ALEC'). The 'minimum argument' that linguistic complexity equalizes across language according to their equal (minimal) amount of work he rejects claiming that ''a great deal of complexity is redundant historical baggage'' (245). The ''maximum argument'' that linguistic complexity ''reaches an upper limit ... due either to the limitation of transmission between generations (learnability), or to the brain's capacity to process'' he denies, asserting ''the widespread phenomenon of bilingualism and multilingualism,'' which ''proves that individual languages do not even begin to exhaust the brain's capacity to learn and process language ... ALEC implies that 'overall complexity' of a language is a meaningful concept that can be non-arbitrarily quantified,'' at best, however, ''the 'overall complexity' of a language can be understood as a vector ... of values'' (246-7). But such vector of values ''will not be amenable to summation, since the parts to be counted in the different subdomains will be of very different natures'' (249).
18. ''An efficiency theory of complexity and related phenomena'' by John A. Hawkins argues that ''metrics of complexity'' should be ''embedded in a larger theory of efficiency...: communication is efficient when [a] message ... is delivered ... in rapid time and with minimal processing effort'' (253). Grammar is not autonomous of performance, as ''widely held in generative theorizing,'' but deeply interconnected with it, so that ''comparing grammars in terms of efficiency in communication as defined ... enables us to model more of the factors that ultimately determine the preferences of performance and grammar, in addition to complexity itself'' (267).
19. In ''Envoi,'' the short final chapter, the editors identify three 'general themes' of the book: authors are ''pushing at an open door'' (268); syntax and morphology are central but syntax particularly as a consequence of ''the Chomskyan revolution,'' to which ''our contributors are not very sympathetic'' (270); and ''there is little reason to expect the real world to contain barriers between'' language and culture (271).
The book is an engaging if unsatisfying argument for the hypothesis of language complexity as an evolving variable. Clearly needed is a better-shared understanding of what 'language complexity' is. Comrie (1992: 210) suggested that this be thought of as what language evolution adds (for example morphophonemic alternation and suppletion). Several authors here offer or allude to such understanding, so perhaps this with other recent work on 'complexity' might be taken with work on language evolution (some of which challenges the traditional dogma that ancient and modern languages do not significantly differ) as groping toward a new language typology based on progress along grammaticalization paths, consistent with the plea of Bybee (2006: 194) ''for the necessity of taking diachrony into account in the formulations of language universals ... language theory must look beyond synchronic generalizations about particular language states to the formational mechanisms that bring linguistic structure into being.''
Bybee, Joan (2006). Language change and universals, Language Universals, Ricardo Mairal and Juana Gil, eds., 179-194. Cambridge: Cambridge University Press. Comrie, Bernard (1992). Before complexity, in Hawkins and Gell-Mann, eds., 193-211. Dahl, Östen (2004). The Growth and Maintenance of Linguistic Complexity. Amsterdam & Philadelphia: Benjamins. Deutscher, Guy (2000). Syntactic Change in Akkadian: the Evolution of Sentential Complementation. Oxford: Oxford University Press. Diessel, Holger and Michael Tomasello (2001). The acquisition of finite complement clauses in English: a usage-based approach to the development of grammatical constructions. Cognitive Linguistics 12.97-141. Everett, Daniel L. (2005). Cultural constraints on grammar and cognition in Pirahã: another look at the design features of human language. Current Anthropology 76.621-46. Everett, Daniel L. (2009). Cultural culture and grammar in Pirahã: a response to some criticisms. Language 85.405-442. Givón, Talmy (1979). On Understanding Grammar. New York: Academic Press. Givón, T. and Masayoshi Shibatani (2009). Syntactic Complexity: Diachrony, Acquisition, Neuro-Cognition, Evolution. Amsterdam & Philadelphia: Benjamins. Hawkins, John A. and Murray Gell-Mann, eds. (1992). The Evolution of Human Languages. Redwood City, CA: Addison-Wesley. Hay, Jennifer and Laurie Bauer (2007). Phoneme inventory size and population size. Language 83.388-400. Hockett, Charles F. (1958). A Course in Modern Linguistics. New York: Macmillan. Jackendoff, Ray (1993). Patterns in the Mind: Language and Human Nature. New York: Basic Books. Kibrik, Aleksandr (1998). Archi (Caucasian - Daghestanian), The Handbook of Morphology, Andrew Spencer and Arnold M. Zwicky, eds., 455-476. Oxford, UK: Blackwell. Kusters, Wouter (2003). Linguistic Complexity: the Influence of Social Change on Verbal Inflection. Utrecht: Netherlands Graduate School of Linguistics. McWhorter, John H. (2001). The world's simplest grammars are creole grammars. Linguistic Typology 5(2/3).125-167. Miestamo, Matti, Kaius Sinnemäki and Fred Karlsson, eds. (2008). Language Complexity: Typology, Contact, Change. Amsterdam: Benjamins. Miller, George A. and S. Isard (1964). Free recall of self-embedded English sentences. Information and Control 7.292-303. Trudgill, Peter (1998). Typology and sociolinguistics: linguistic structure, social structure and explanatory comparative dialectology. Folia Linguistica 31.349-60.
ABOUT THE REVIEWER:
ABOUT THE REVIEWER:
Grover Hudson taught phonology, historical linguistics, and Ethiopian
linguistics including Amharic language at Michigan State University. He is
author of a comparative dictionary of Highland East Cushitic languages, an
introductory linguistics textbook, with Anbessa Teferra a recent book on
Amharic, and articles on phonology and Ethiopian descriptive and historical