weng_1651
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48
wengxml-als.cls (1994/07/13 v1.2u Standard LaTeX document class)
3-13-2010
:1282
WENG
weng_1651
Dispatch: 3-13-2010
CE: N/A
Journal
MSP No.
No. of pages: 4
PE: Yana
World Englishes, Vol. 29, No. 2, pp. 321–324, 2010.
0883-2919
RESPONSE to Croft∗ SALIKOKO S. MUFWENE∗
William Croft’s (WC’s) review of my Language Evolution: Contact, Competition and Change (LE) is constructive, perceptive, and so well balanced that I would be remiss to respond to it defensively, especially when it is intended to establish a dialog between two competing accounts of language evolution that are fundamentally inspired by evolutionary biology. However, I must clarify that this book is not largely an English version of ´ ´ Cr´eoles, Ecologie Sociale, Evolution Linguistique (2005), although it elaborates many of the same topics discussed in the latter. Unlike the former, LE includes a chapter on pattern emergence at the population level (explaining how communal norms emerge), a chapter on grammaticization, one on competition and selection (clarifying that competition is a misnomer for features being ranked differently by their users, see below), and a richer discussion of how globalization becomes an important ecological factor in language evolution (including language spread, speciation, and in some cases loss). What WC reports correctly is that in the three books since The Ecology of Language Evolution (Mufwene 2001) I continue to develop an account of language evolution that is inspired by population genetics and macroecology. The former has prompted me to pay more attention than traditionally in linguistics to the population/species aspect of languages as extrapolations from idiolects, the latter being the counterparts of biological organisms. This is significant because the actions that bring about change lie in isolated but repeated communicative activities of individual speakers, thus raising the question of how some but not all the changes experienced by individual speakers spread into changes at the communal level, at which linguists discuss language evolution. Macroecology has inspired me to think harder about the actuation of change, suggesting in fact that, like biological evolution, language evolution is generally externally motivated; there are no internally motivated changes. All the changes that languages undergo reflect our linguistic behaviors under particular ecological pressures, including the influences of language varieties we already speak and of the particular individuals we hope to communicate successfully with. The latter factor prompts us to accommodate one another in a variety of ways. Consistent with the construction-grammar approach practiced by WC, grammatical patterns emerge, not because speakers plan them, but as outcomes of repeated ways of selecting and producing particular features (sounds, words, and even phrases). As explained in ELE, there are different kinds of species, even in biology, and they do not all display similar ontogenetic or evolutionary properties (see below). WC is very perceptive in noting the apparent contradiction in my analogizing linguistic features qua units of selection to genes in biology and yet denying that the ∗ Department of Linguistics, University of Chicago, 1010 East 59th Street, Chicago, IL 60637, USA. E-mail:
[email protected] C 2010 The Author(s). Journal compilation C 2010 Blackwell Publishing Ltd., 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.
weng_1651
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48
wengxml-als.cls (1994/07/13 v1.2u Standard LaTeX document class)
322
3-13-2010
:1282
Response to croft
genotype-vs-phenotype opposition assumed in biology does not apply to languages. As explained in both ELE and LE, unlike biological organisms, which are conceived wholesale, idiolects emerge gradually and incrementally, through ‘language acquisition’. Their properties are determined by what ‘interactors’ (speakers or signers) perceive as linguistically significant in their communicative ecology and integrate in their communicative patterns. I actually stated in ELE that languages are species in their own right. There are peculiarities of their evolutions (including speed of change) that are consequences of their ontogenetic properties which are different from those of animal species, including transmission/replication with modification, polyploidy (which allows multiple influences from various speakers on the same individual – an important factor in shaping idiolects), and the fact that the feature composition and structure of an idiolect change so many times in the life of a speaker. These are indeed considerations that led me to analogize languages originally to parasites and later, on the correct advice of William Kretzschmar (pers. comm. 2003), to viruses. As organisms, idiolects are indeed more Lamarckian than Darwinian, since they frequently change their genetic makeups while they adapt to their different hosts, on whose life style their vitality depends. This is precisely the reason why I rejected Richard Dawkins’ (1976/1989) memetics as not being the right application of the biological model to cultural evolution, language being an instantiation of culture. In this kind of approach one is inevitably confronted with the question of what the units of selection are, which presupposes another, namely, what are the elements that are in competition? Linguistics has made it very easy for us to address the latter, as various units have been posited in various modules, namely, sounds, morphemes, words, and why not phrases. Now that linguists are very much interested in language vitality (roughly, which ones prevail at the expense of others, and where?), we can add dialects or languages. (Interestingly, attitudes to idiolects are rather co-operative, fostering variation within a language variety.) It is of course through utterances (units of communication in our regular interactions) that we perceive these particular units and subject language varieties to competition, as we usually intend to speak one at a time. However, the question is whether we can stipulate utterances as replicators. Although speakers often repeat entire phrases they have heard before (quite typically in idiomatic communication), we often also copy, selectively indeed, words or particular pronunciations perceived in other speakers’ utterances. To be consistent with how Richard Dawkins defined memes, one would have to acknowledge that there are different kinds of replicators, which vary in length and complexity. Linguistic feature and lingueme appear to be more suitable blanket designations than utterance for the different units of selection. WC is correct in remarking that he and I have applied inspiration from evolutionary biology differently. While I argued in ELE that linguistics can also contribute in its own way to theories of evolution, I have resisted assuming that there is already “a generalized theory of evolutionary change” out there, with its “general analysis of selection,” that applies uniformly to all species-like phenomena. I think we still need to articulate clearly how language evolution resembles what particular kind of biological evolution and how it differs from it. This is part of what I do in fact in chapter 2 of LE (and in various parts of the previous books). I will now focus on various details of WC’s informative and constructive review. Do I really invoke the “invisible hand” differently from Rudi Keller (1994)? Patterns emerge in a communal language because different speakers do more or less the same C 2010
C 2010 Blackwell Publishing Ltd. The Author(s). Journal compilation
weng_1651
wengxml-als.cls (1994/07/13 v1.2u Standard LaTeX document class)
3-13-2010
:1282
1
Response to croft
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48
thing, analogously to pedestrians walking the same course across a lawn and ultimately developing an unplanned path. Although many do it because it is just more convenient for them to walk that particular course (thus they innovate independently of each other), there are also many others who do it because they have seen others do it without being penalized. Eventually the new path becomes practical for many or all. The emergent path is thus an evolutionary phenomenon. There is actually a more original conception of the “invisible hand” as conceived of in economics by Smith (1776), namely, everybody is behaving in a particular way that is beneficial to them. Theoretically such individualistic behavior could create conflicting trends in a particular population. However, one notices later on that the individual behaviors have converged into communal patterns, or norms in language, that are beneficial to the overall population. In their usually dyadic and triadic interactions, speakers typically focus on being understood, maximizing success in manipulating others but not anticipating the emergence of norms. How we wind up converging toward some dominant patterns at the expense of others, subject to various ecological pressures, is still far from being clear. I invoke the “invisible hand” in all the above ways, drawing attention again to population-internal dynamics that have not been explained yet but generate new norms and sometimes new language varieties, and in some other cases drive some language varieties to extinction. While the focus of studies of language evolution has been at the population level, it is important to pay attention to the practices of individual speakers, because speakers are the actual agents of change. The “invisible hand” is of course not an evolutionary theory; it is a mechanism that produces evolution. It encapsulates another dimension of selection, though this is executed by the same individuals who continually (re)shape their idiolects selectively from materials available in their communal feature pools and, in multilingual settings, repeatedly select a particular language for communication. I think it is important to think about this particular aspect of how evolution proceeds when we know that language behavior is not like a sport with explicit rules of engagement toward common goals. The goal of linguistic communication is getting one’s message across and not developing a particular communal norm. Yet, the latter emerges. How? This is indeed an important question that an adequate evolutionary theory should not dodge. WC is right. While I find genetic linguistics still wedded to some 19th-century racerelated tradition of language purity that treats contact-based language varieties as genetic anomalies, I do not accuse genetic linguists of being racist. The most revolutionary of us still carry some legacy of what we stand against! On the other hand, I must clarify that the regularities of sound and other form correspondences on which the comparative method relies is independent of the origins of the correspondences. As Meillet (1900) and Tremblay (2005) in particular point out, the related languages may owe the shared features to a common ancestor, may have borrowed them from the same language (subject to ecological pressures that yielded the relevant regular adaptations), or may have innovated them under similar conditions. There is no particular excuse why uniparental genetic relatedness should have been considered as the norm when the Romance languages, which present the most documentation for the paradigm, suggest hybridization as the normal mechanism toward language speciation. Regarding language endangerment and loss, I doubt that the controversy can be reduced to the simple fact that more languages are dying now than in earlier stages of human C 2010
C 2010 Blackwell Publishing Ltd. The Author(s). Journal compilation
323
weng_1651
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48
wengxml-als.cls (1994/07/13 v1.2u Standard LaTeX document class)
324
3-13-2010
:1282
Response to croft
history. I think the same case could have been made during the loss of Celtic languages in Western Europe during the expansion of especially Vulgar Latin and, later, English as vernaculars. While language advocacy is laudable citizenship, linguists should develop more adequate and informative scholarship, on the model of environmentalists in regard to endangered species, about language loss. They should be more realistic about what they can lead the relevant populations to expect from their interventions. Just like WC, I take the following distinction seriously: (1) “language documentation” is part of a museum culture and is most useful to linguistics; (2) “language preservation” is a misnomer for products such as dictionaries, grammars, and other books that are not the language itself but give us glimpses into particular phases in the history of its use (as in the case of Hittite and Latin, for instance); and (3) “language maintenance,” dealing with actual usage, requires much more than linguistics as a profession can provide. In LE, I bring to bear many interesting cases from Africa that most language advocates have not talked about. I also like WC’s invocation of urbanization. Mufwene and Vigouroux (2008) show that this has not proceeded uniformly in Africa and in the West. The pressure toward monolingualism is less intense in Black African than in Western cities. Africa may be urbanizing very rapidly now, but with more destitute neighborhoods than the West. Overall, the rate of actual urbanization is still much lower in Africa than in the West, as I intend to show in later work. I am sure WC and I will have more opportunities to trade notes on this subject too. REFERENCES Dawkins, Richard (1976/1989) The Selfish Gene (new edition). New York: Oxford University Press. Keller, Rudi (1994) On Language Change: The Invisible Hand in Language. London: Routledge. (Translation and expansion of Sprachwandel, 1990.) Meillet, Antoine (1900) Note sure une difficult´e g´en´erale de la grammaire compar´ee. Reprinted in Linguistique historique et linguistique g´en´erale (pp. 36–43). Geneva: Slaktine. Mufwene, Salikoko (2001) The Ecology of Language Evolution. Cambridge: Cambridge University Press. Mufwene, Salikoko S. & Vigouroux, C´ecile B. (2008) Colonization, globalization and language vitality in Africa: an introduction. In C´ecile B. Vigouroux & Salikoko S. Mufwene (eds.), Globalization and Language Vitality: Perspectives from Africa (pp. 1–31). London: Continuum. Tremblay, Xavier (2005) Grammaire compar´ee et grammaire historique: quelle r´ealit´e est reconstruite par la grammaire compar´ee? In G´erard Fussman, Jeans Kellens, Henri-Paul Francfort, and Xavier Tremblay (eds.), Aryas, Aryens et Iraniens en Asie Centrale (pp. 33–180). Paris: Edition-Diffusion de Boaccard. Smith, Adam (1776) The Wealth of Nations, vol. 1. Oxford: Clarendon Press. (Received 28 September 2009.)
C 2010
C 2010 Blackwell Publishing Ltd. The Author(s). Journal compilation
weng_1651
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48
wengxml-als.cls (1994/07/13 v1.2u Standard LaTeX document class)
3-13-2010
:1282
