Review of Language Complexity as an Evolving Variable
|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
Connie de Vos, Max Planck Institute for Psycholinguistics
This book is the end-result of a workshop held at the Max Planck Institute for
Evolutionary Anthropology in Leipzig in April 2007. Although the workshop set
out to dispute the linguistic axiom that all languages are equally complex, the
editors mention in the final chapter (''Envoi'') that the meeting attracted only
those who are actively challenging that view. The book provides a rich view on
the range of definitions, linguistic phenomena, and methods concerning the issue
of linguistic complexity. Methods vary from statistical analyses of large sets
of spontaneous conversations to the detailed analysis of a single creole
language. Most strikingly, the papers in this book show that linguistic
complexity may vary between languages, within and across speakers of a language,
over time at the individual level (even past puberty) and language level.
Multiple hypotheses are put forward on the kinds of factors which may impact
language complexity. The papers in this book give a representative overview of
such factors: individual factors, and literacy, but also sociolinguistic factors
such as language contact, percentage of L2 learners, time depth, and community
The first paper by Geoffrey Sampson explains why the axiom that all languages
are equally complex has been so attractive to various schools of linguistics,
and how new empirical work challenges it. Central to the challenge based on
linguistic data is to disprove a mechanism of trade-off between various domains
of linguistic structure which would make languages equally complex in the end.
David Gil (chapter 2) argues that even the least complex languages are
functionally sufficient for the most complicated tasks; that language complexity
evolves from system-internal processes, rather than in response to a functional
Walter Bisang's paper argues that while some languages may exhibit complexity on
the morphosyntactic level, this is compensated for through pragmatic inference
in languages with less morphosyntactic marking (e.g. languages of South-East
Asia). According to Bisang, this should be considered as hidden complexity.
In chapter 4, Östen Dahl compares the grammars of Elfdalian and Swedish - two
closely related languages - with respect to phonological, morphological, and
syntactic complexity. Although there have long been presumed to be trade-offs
between domains of linguistic structure such that overall language complexity is
equal, Dahl's data show that there is no apparent trade-off.
In chapter 5, Benedikt Szmrecsanyi and Bernd Kortmann present a frequency-based
analysis of English used spontaneously in different societies: English spoken as
a native language by high-contact and low contact communities, English spoken as
a second language, and English-based pidgins and creoles. Linguistic complexity
is negatively correlated with contact.
In chapter 6, Matti Miestamo argues for the use of Greenbergian implicational
hierarchies in determining whether there are trade-offs between linguistic
domains which would support an equi-complexity hypothesis. His stratified sample
of languages shows that trade-offs do not exist between agreement and case
hierarchies, but there is significant trade-off between the complexities of
copular and verbal marking in stative predicates. Unfortunately, the domains for
which implicational hierarchies exist do not exhaustively describe the
linguistic complexity of any particular language, which limits the use of
implicational hierarchies for research on complexity.
Peter Trudgill (chapter 7) argues that societal type influences linguistic
structure such that more contact, looser networks, and a larger community result
in simplification (i.e. regularization, transparency, reduction of redundancy)
because of L2 learner preferences.
In chapter 8, Johanna Nichols argues in favor of sociolinguistic factors
influencing language structure based on the grammars of sixty-eight languages.
However, she also points out there are many remaining questions on how these
factors play out in combination, and over time. Population size for instance may
fluctuate rapidly, while the structural properties of the language may need time
Kaius Sinnemäki (chapter 9) shows an impact of population size on core argument
marking such that larger languages are more likely to have only a single
strategy to distinguish 'who did what to whom', while smaller languages may
deviate more easily from this one-meaning-one-form principle.
In chapter 10 John McWhorter shows how the creole language Saramaccan has
developed a new information marker, which cannot be traced back to the source
Utz Maas (chapter 11) explains how literacy increases language complexity since
a register for usage is added with respect to oral languages.
Ngoni Chipere's work (chapter 12) discusses individual differences in the
processing of complex linguistic structures. Training studies indicate that
these differences are the result of varying grammatical competence rather than
differences in working memory.
In chapter 13 Fred Karlsson discusses the origin and maintenance of clausal
embedding complexity in Standard Average European languages under the influence
of Latin writing.
In chapter 14 Ljiljana Progovac argues for the gradual evolution of syntax based
on the structure of present-day grammar.
In addition to the papers, chapter 15 presents an email-based interview by
Geoffrey Sampson with Dan Everett: the linguist whose controversial work
involves Pirahã, one of the least complex languages known to date. According to
Everett, the relatively simple language Pirahã fits the cultural needs of the
In chapter 16 Eugénie Stapert looks at the complexity of mental verb
constructions in both Pirahã and English language acquisition.
Guy Deutscher (chapter 17) addresses theoretical and practical problems with
Chapter 18, by John A. Hawkins, describes various efficiency factors which
optimize communication and argues for comparing grammars with respect to
efficiency rather than complexity.
Chapter 19 is a concluding chapter by the editors.
An axiom challenged
Geoffrey Sampson starts off by questioning why the axiom of equal complexity
across languages has dominated linguistics for most of the twentieth century.
For Chomskyan linguists, equal complexity follows naturally from the hypothesis
of Universal Grammar. That is, since Universal Grammar is the same across the
globe and innate to all peoples each language has an equal amount of structural
complexity. There is thus an a priori reason why linguistic complexity should
exist as a constant. The descriptivist's soul is a bit more complicated. The
main reason why they have been supportive of the constant complexity axiom seems
an ethical one: ''all men are brothers''. Note that this line of thinking invokes
in fact another assumption, namely that linguistic complexity may impact on
cognitive capacity or degree of civilization which is subsequently seen as a
(moral) judgment. Consequently, linguists in the descriptivist tradition have
been in favor of the idea that simplicity in one domain of linguistic
organization makes up for complexity in another domain of linguistic structure.
These trade-offs cause language complexity to be in a state of equilibrium.
Sampson's contribution is a valuable reflection on a changing linguistic paradigm.
Variation in Language complexity
Challenges to the idea of equal complexity across languages come from various
empirical works. First, there are typological descriptions of fully functional,
non-creole languages which nevertheless exhibit little complexity in any overt
domain. This includes work by David Gil on Riau Indonesian (see also chapter 2),
as well as Dan Everett's work on Pirahã (see the interview in chapter 15).
Chapters 2, 3, 4, 6, and 8 show that a trade-off of complexity between overtly
marked linguistic domains is not generally supported by the data. What is more,
diachronic studies have shown that languages may develop more syntactic
complexity over time (see for example chapters 11, 13, and 14).
In contrast to the above studies that focus on complexity as absolute, system
complexity, the chapters by Szmrecsanyi & Kortmann (chapter 5), Peter Trudgill
(chapter 7), and Kaius Sinnemäki (chapter 9) look at complexity from an L2
learners' perspective. They thus define complexity as degree of difficulty of
acquisition. These studies show that societal variables such as community size,
network density and amount of contact may influence a language's structure.
Comprehension studies by Ngoni Chipere (see chapter 12) show that there are
individual differences within a community in the processing of complex
sentences. Surprisingly, these studies show that L2 learners are in fact more
accurate than native speakers in such cases. In fact, Chipere's studies do not
only challenge the idea of constant linguistic complexity across individuals,
but at the same time they make one question the robustness of the Critical
Period in the syntactic domain.
This book presents papers by leading researchers in the field of linguistic
complexity, most of whom have published monographs or edited volumes on the
topic before. (See for instance Dahl 2004; Miestamo, Sinnemäki & Karlsson 2008;
Trudgill forthcoming.) This edited volume can therefore serve as a first
introduction to the issue of language complexity, but also as an update for
those who are more familiar with the research. In general, the papers are
well-written and although definitions vary widely, they are made explicit in
each chapter. In fact, the variety of ways in which linguistic complexity is
investigated makes the case against equal complexity across languages much more
compelling than any of the individual work would by itself.
The question whether linguistic complexity (language) may be a gradual
phenomenon also plays a role in sign linguistics. For instance, home sign -
communication by deaf children without a conventional language model - displays
evidence of linguistic structure (Goldin-Meadow 2003). Moreover, the comparison
of different cohorts of a well-documented, young signed language in Nicaragua
showed linguistic conventionalisation patterns over just a few decades (Senghas,
Kita, & Ozyurek 2004). Recently, so-called village sign languages - young
signed languages of relatively isolated, small communities with a high
percentage of hearing non-native signers - are receiving more attention in sign
linguistics (Meir, Sandler, Padden, & Aronoff in press). The present book,
however, contains neither work nor references to research on signed languages.
This is a pity, especially because many of these languages' histories can be
traced. As such village sign languages would form a perfect testing ground for
theories which implicate sociolinguistic factors in the development of language
Although Sampson et al.'s edited book was published in a series called Studies
on the Evolution of Language, and the book title makes reference to evolution,
there are virtually no remarks on what makes language change like Darwinian
evolution, or how the data presented in this book would contribute to research
on the emergence and evolution of language. I feel this is a missed opportunity.
The idea of varying complexity has far-reaching consequences for biologists or
geneticists interested in the phenomenon of the emergence and evolution of
language in our species. It changes our view on what the human linguistic
phenotype really looks like. In fact, it may well be that the tremendous
flexibility of what languages can be like makes human communication unique.
Dahl, Ö. (2004). The growth and maintenance of linguistic complexity. Amsterdam:
Goldin-Meadow, S. (2003). The resilience of language. What gesture creation in
deaf children can tell us about how all children learn language. New York, NY:
Meir, I., Sandler, W., Padden, C., & Aronoff, M. (in press). Emerging Sign
Languages. In M. Marschark & P. E. Spencer (Eds.), Oxford Handbook of Deaf
Studies, Language and Education, Volume 2. Oxford: Oxford University Press.
Miestamo, M., Sinnemäki, K., & Karlsson, F. (Eds.). (2008). Language complexity:
typology, contact, change. Amsterdam: Benjamins.
Senghas, A., Kita, S., & Ozyurek, A. (2004). Children creating core properties
of language: Evidence from an emerging sign language in Nicaragua. Science,
Trudgill, P. (forthcoming). Language in contact and isolation: sociolinguistic
typology and linguistic complexity. Oxford: Oxford University Press.
| ABOUT THE REVIEWER:
ABOUT THE REVIEWER
Connie de Vos is a PhD candidate of the Language & Cognition group at the
Max Planck Institute for Psycholinguistics. She received her MA in
Linguistics at Radboud University, with a thesis on the interaction of
affective and linguistic functions of eyebrows in Sign Language of the
Netherlands. Currently, she is writing a dissertation on Kata Kolok, a
village-based sign language of North-Bali. Her work focuses on pointing,
the use of sign space to talk about spatial relations, and processes of