Zaire, Robins & Uhlenbeck, WCMC, speciation, Australia, Mexico, cultural diversity, China, Indonesia, linguistic diversity, Ethiopia, SOUTHWEST JOURNAL, Tanzania, Schleicher, allopatric speciation, natural world, biological species concept, Cameroon, Genetic diversity, linguistics, Harvard University Press, New York, plant species, Language death, endangered languages, SOUTHWEST JOURNAL OF LINGUISTICS, Ruhlen, endemic species, Papua New Guinea, Belknap Press, biological diversity, Philippines, Island countries, Tropical countries, continental countries, Origin of species, linguistic science, biology, William Dwight Whitney, natural selection, Sir William Jones, Junggrammatische Richtung, Jacob Grimm, Darwin, George Wright Society, Historical linguistics, evolutionary biology, vertebrate species, Otto Jespersen, August Schleicher, levels of biological organization, JOHN A. ENDLER, EDWARD O. WILSON, Oxford University Press, RUSSELL A. MITIERMEIER, Summer Institute of Linguistics, University of Chicago Press, KENTON R. MILLER, Indiana University Press, species concepts, species diversity, Stanford University Press, COLLINGE, World Conservation Monitoring Centre, WALTER V. REID, Vance Martin, ecosystem level, Cambridge University Press, ecosystem levels, Luigi Luca Cavalli-Sforza, Solomon Islands, Joseph Greenberg, global biodiversity, species concept, Conservation International, phylogenetic species concept, species richness
LOSING SPECIES, LOSING LANGUAGES: CONNECTIONS BETWEEN BIOLOGICAL AND LINGUISTIC DIVERSITY DAVID HARMON The George Wright Society 1. EMERGENCE OF EVOLUTIONARY THOUGHT IN BIOLOGY AND LINGUISTICS. On the face of things, it is remarkable that two fields of inquiry so seemingly different as biology and linguistics should have followed such similar paths toward the same fundamental tenet: that the process of change is evolutionary, with its course to the present traceable to ancestors. This conclusion, though second nature to us now, shattered the intellectual world of Western Europe in the 19th century. It displaced an insular, dogmatic outlook in which many baffling questions--such as the reason for the great plenitude of languages and species in the world--could be explained only by convoluted reference to a largely static, largely predetermined grand order. The Great Chain of Being, current since the time of Plato, is the bestknown conceptualization of this world-view, and it or some variant of it held sway until the last decades of the 18th century. Then, as Ruhlen (1991:25-28) recounts, thinkers in a number of fields, including linguistics, began groping toward an embryonic conception of evolution. This was all long before Darwin, of course. His great achievements in Origin of species(published in 1859) were to crystallize the concept of evolution and then elaborate it with the theory of natural selection. In so doing he managed to consolidate the strains of evolutionary thought then current in such fields as geology, paleontology, and biology itself. Historical linguistics, it must be said, needed no such consolidation, as its basic comparative methods were already well-established by the work of Rask, Bopp, Jacob Grimm, and others (see Pedersen 1962:248-62; Jespersen 1922:22 ff.). These men all were following the methods that had been signalized by Sir William Jones in his famous Asiatick Society address of 1786. Nonetheless, the astounding speed with which Darwin's views gained ascendancy set something of an imprimatur on what had been happening in linguistics. The relevance of Darwin's insights was grasped immediately by such linguists as August Schleicher. Four years after the publication of Origin, Schleicher came out with a pamphlet called Darwinism tested by the science opanguage. ·What Darwin now maintains with regard to the variation of the species ... has been long
90 SOUTHWEST Journal of Linguistics
, VOLUME 15, NUMBERS 1 & 2 (1996) and generally recognised in its application to the organisms of speech', he observed dryly. 'To trace the development of new forms from anterior ones is much easier, and can be executed on a larger scale, in the field of speech than in the organisms of plants and animals. ... The kinship of the different languages may consequently serve ... as a paradigmatic illustration of the origin of species ...' (quoted in Law 1990:819). Darwin himself recognized the basic affinities between evolutionary biology and historical linguistics. He drew the parallel in The descent of man (published in 1871), likening the process of forming species and languages (see quotes in Ruhlen 1991:25 and in Ruhlen 1994:136-37), as well as in his Origin, where he declared that, if we could but properly classify all the languages of the world, both living and extinct, along with 'all intermediate and slowly changing dialects', we would of necessity arrive at 'a perfect pedigree of mankind' (quoted in Ruhlen 1994:160). These sentiments were echoed by Schleicher, who claimed that the classification of humans into races should not be based on physical appearance or any other external criteria, 'as they are by no means constant, but rather on language, because this is a thoroughly constant criterion. This alone would give a perfectly natural classification system.' Otto Jespersen, who provides this paraphrase, goes on to demolish the idea, asking whether we are `to reckon the Basque's son, who speaks nothing but French (or Spanish), as belonging to a different race from his father?' (1922:75). Indeed, a counter-reaction against taking the language--species parallel too far had set in very shortly after Schleicher's death in 1868. Among the things the 'Young Grammarians'(Junggrammatische Richtung) revolted against was Schleicher's assertion that each language is 'a natural organism and its study a natural science' (Col li nge 1990b:880-81). So did other linguists outside Germany, such as William Dwight Whitney, who wrote in 1875 that 'physical science on the one side, and psychology on the other, are striving to take possession of linguistic science, which in truth belongs to neither.' Vivien Law, who provides this quote, goes on to say that this plea for the autonomy of linguistics was to echo on through the twentieth century' (1990:820). She is quite right. Here is Jespersen again: The distinctive feature of the science of language as conceived nowadays is its historical character: a language or a word is no longer taken as something given once for all, but as a result of previous development and at the same as the starting-point for subsequent development. This manner of viewing languages constitutes a decisive improvement on the way in which languages were dealt with in previous centuries, and it suffices to mention such words as 'evolution' and 'Darwinism' to show that linguistic research has in this respect been in full accordance with tendencies observed in many other branches of scientific work during the last hundred years. Still, it cannot be said that students of language have always and to the fullest extent made it clear to themselves what is the real essence of a language. Too often expres-
LOSING SPECIES, LOSING LANGUAGES
sions are used which are nothing but metaphors that obscure the real facts of the matter. Language is frequently spoken of as a 'living organism'; we hear of the 'life' of languages, of the 'birth' of languages and of the 'death' of old languages, and the implication, though not always realized, is that a language is a living thing, something analogous to an animal or a plant. Yet a language evidently has no separate existence in the same way a dog or a beech has, but is nothing but a function of certain living human beings. Language is activity, purposeful activity, and we should never lose sight of the speaking individuals and of their purpose in acting in this particular way (1922:7). Collinge, reviewing all this in a recent encyclopedia article, remarks, pace Schleicher and the other 'linguistic Darwinists of the last century', that 'languages and their forms are not some kind of natural organism' (1990b:899). This conclusion, while valid enough, is really something of a straw man. No biologist would ever equate a language with an individual member of a species, for the simple reason that humans can choose which language or languages to use, whereas volition does not play a role in determining which species an organism belongs to. Try as it might, a toucan can never become a jaguar--though the whole point is that it is quite beyond the toucan's capacity to try. No non-human species
can consciously direct its own collective future, while the course of a given language is, ultimately, a matter of choices made by both speakers and non-speakers. To put it another way, languages change by cultural selection, species by natural selection. A more pointed criticism conies from Nichols (1992). Reviewing different senses of the term 'evolution' as they apply to linguistics, she says: No evidence of anything like speciation has been found in this or any other typological work. Although linguistics has no analog to the biological notion of species, it is safe to say, informally speaking, that languages and linguistic lineages are related to each other as individuals or kin groups of a biological species are, not as species in a genus' (277). Again, one must at least partly agree with her, for undoubtedly no language has ever diverged from all others to the point that it is absolutely impenetrable, is not capable of being learned by outsiders, is--to put it in biological terms--'reproductively isolated' from other languages. Such an occurrence would indeed make for an entirely different `species' of language in the strictest sense of the word. But I strongly disagree that there is no linguistic ANALOGUE to biological species; to my mind, an analogy is a comparison where the broad outlines run parallel while the particulars may not. I think there is a convincing analogy to be drawn between a species and a language. More importantly, I will try to show that this analogy is far from being one ofJespersen's 'metaphors that obscure the real facts of the matter'; rather, it is useful to a practical consideration of the nature and extent of linguistic diversity.
92 SOUTHWEST JOURNAL OF LINGUISTICS, VOLUME 15, NUMBERS I & 2 (1996) 2. SPECIATION AND LANGUAGE GENESIS. So a language is not the same thing as a species. Yet languages and species ARE similar in many powerful ways. This ought not be terribly surprising, since both are, in the most abstract sense, expressions of cohesion for groups of individuals. They are givers of identity, bestowers of distinctiveness. This broad, abstract similarity is valuable in itself because it gives us a basis on which to compare diversity in language and nature. But there are also more concrete similarities, beginning with the problem of how to define species and languages themselves. The puzzle of how to distinguish languages from dialects, and indeed languages from other closely related languages, is of long standing. To some linguists, it is well-nigh intractable. Durrell, for example, asserts that there are `no satisfactory objective criteria' for determining what is a language and what a dialect. As he rightly points out, the layperson's criterion, mutual intelligibility, is by itself not definitive, since there are many mutually intelligible autonomous languages and, conversely, mutually unintelligible dialects of the same language. 'In fact, in the conventional designation of a particular geographically determined variety of a "dialect" of a particular "language", our definition of "language" is still based on geopolitical, ethnic or cultural rather than linguistic criteria ...' (Durrell 1990:921). Vexed by the difficulty of distinguishing languages from dialects, of separating discrete from continuous variation in language, it is easy for linguists to look at biology as being far more precise than their own field: The language/dialect distinction in linguistics parallels the species/variety distinction in biology. Whereas a species is usually thought of as a group of plants or animals that can interbreed and produce viable progeny, a language is a group of dialects among which there is mutual intelligibility. ... But the analogy is not perfect, since the criterion of mutual intelligibility is less well defined than that of interbreeding. Mutual intelligibility ranges from close to 0 percent to close to 100 percent; interbreeding is usually an all-or-nothing proposition. ... IT1here are many gray areas in deciding what to call a language and what to call a dialect. Biological speciation is much more clear-cut (Ruhlen 1991:21). In truth, this conclusion is much too generous to biology. Although there is a strong case to be made for the mode of speciation alluded to above, the issue is far from settled. Proponents of the biological species concept (of which more below) argue that the predominant (or even the only) way species form is by 'allopatric speciation', in which a population becomes geographically isolated from the rest of the species so that it eventually becomes reproductively isolated as well and can no longer interbreed (see Mayr 1963:451-80). However, there is a strong undercurrent of dissent to this viewpoint, centered among certain botanists, genetic biologists, and taxonomists. They point out that there are plenty of cases of `sympatric speciation' in which geographical isolation is not a prerequisite for the formation of a new species (see the papers in Otte & Endler 1989). Edward 0. Wilson, widely
LOSING SPECIES, LOSING LANGUAGES
considered one of the most influential biologists working today and recognized as a leader in the movement to preserve biodiversity, comes down squarely in favor of the primacy of allopatric speciation, but acknowledges that 'it is supplemented in nature by a rich medley of other modes of speciation' (Wilson 1992:69). Speciation is not, therefore, quite so clear-cut. What's more, biologists have just as much trouble defining species as linguists do language. Hard though it may be to believe, the very validity of the species concept is still up for debate: What is a species? This simple question has troubled biologists for inore than two centuries. Although accepted so widely as a 'nature, basic or fundamental unit, many conflicting definitions of species have been coined, and agreement is still lacking. The range of definitions reflects, to a large degree, the differing interests and differing theories of individual scientists about the origin of diversity itself--literally from Genesis to Darwin and DNA (WCMC 1992:13). There are two main competing ideas about what constitutes a species. One, called the 'biological species concept', is what most of us think of when we think of a species: The basic idea of a biological species is that of a 'pool' of genes available for re-combination through sexual reproduction, but not with genes belonging to other gene pools, from which they are `protected' by a variety of recognition and isolation mechanisms (behavioural, physiological, genetical, etc.). Thus the biological species to which a given individual belongs is determined by the limit of the populations with which it interbreeds, or potentially interbreeds (WCMC 1992:14). This concept is 'probably the most widely accepted view of the species held by biologists today' (WCMC 1992:14) It is not beyond controversy, however. Templeton (1989) and Cracraft (1989), among many others, argue forcefully that the biological species concept, while intuitively attractive, is simply inapplicable to great numbers of organisms. They make two basic objections: (1) the concept simply ignores the large class of organisms that reproduce asexually or which self-mate, and (2) there are so many exceptions to the concept, even among sexually reproducing organisms, that it loses much of its validity. These exceptions involve evidence that there is considerably more gene flow BETWEEN SPECIES than should take place if they were truly reproductively isolated, as the biological species concept holds. In other words, On the 'fringes' of numerous species there is a lot of genetic exchange going on (see MacArthur 1972:71-73)--rather like the linguistic exchange that goes on between dialects along the margin of their language's ranges. Such naturally occurring hybridization is commonplace among plants (botanists have a special term, the 'syngameon', to denote aggregations of plant species that exhibit gene exchange); now, with the advent of genetic fingerprinting techniques, there is evidence that it takes place even among mammals (Templeton 1989:11, 1991). This violates the most basic tenet of the biological species concept: the separateness of gene pools (WCMC 1992:14).
94 SOUTIIWEST JOURNAL OF LINGUISTICS, VOLUME 15, NUMBERS 1 & 2 (1996) Those who criticize the biological species concept have proposed alternatives that either emphasize (1) mechanisms of reproductive cohesion rather than isolation (Templeton 1989:12-24) or (2) differentiation signified by historical relationships between whole taxa rather than current blood relationships of individuals. The latter alternative, standing in opposition to the biological species concept, is known as the `phylogenetic species concept'. It views species as irreducible clusters of organisms `diagnosably distinct from other such clusters, and within which there is a parental pattern of ancestry and descent' (Cracraft 1989:34-35). The problem with the phylogenetic species concept is that in the search for 'diagnosable distinctions' one could lose sight of the forest for the trees and, 'by reductio ad absurdum, every population, stage, morph [i.e. variant], or even individual organism could be elevated to separate species status' (WCMC 1992:15). It is not to our purpose to go into this debate in any further detail. What has been said so far points to four important things to keep in mind when comparing species and languages, and the larger process of speciation with language genesis. First, those who would define species must face exactly the same kind of problem as those who would define languages: that of separating discrete from continuous variation at the margins, where species blend into species and languages blend into languages (or dialects into dialects). Isn't it striking that the two defining criteria at issue--reproductive isolation for species, and mutual intelligibility for languages and dialects--are both so intuitively appealing, and work quite well in so many instances, yet are both faulty in many instances more, enough so that neither is alone sufficient to make a clean definition? Second, the explanation of geographic and temporal variation in species diversity is one of the central problems of biology. It has also proved to be one of the most intractable. The problem has generated an enormous amount of literature in which many different hypotheses have been proposed to attempt to account fOr it; TuIsn HYPOTHESES OIMiN OPERATE AT DIFFERENT LEVELS OF EXPLANATION and much confusion has arisen as a result' (WCMC 1992:46; emphasis added). The same can be said for discussions of linguistic diversity. This problem of differing levels of explanation will be examined below. Third, the idea of allopatric speciation being supplemented by other modes of speciation is mirrored in the processes of language genesis. If we go ahead and accept the primacy of the biological species concept and allopatric speciation-- keeping always in mind that it is augmented in the real world by Wilson's 'rich medley' of other modes, then the most important aspects of speciation are similar to those governing the genesis of new languages. Like speciation, the primary mechanism of language genesis is isolation--in this case, communicative rather than reproductive isolation--which follows geographic separation. The result is that 'forms of speech that are mutually intelligible (i.e. dialects of one language)
LOSING SPECIES, LOSING LANGUAGES
will gradually drift apart, provided only that they are isolated from each other for a sufficient period of time'--roughly 500-1,000 years, according to Ruhlen (1991:6). To borrow the biological term, what is described here is `allopatric language genesis'. It is apparent that geographic isolation plays a key role in allowing the rise of differentiated languages. Indeed, this is the premise undergirding the human origination--migration debate, in which linguistics has become enmeshed (and of which more below). However, as in biology, the allopatric mode doesn't explain all language distributions. Aboriginal Australia
, for one, provides notable examples of sympatric language genesis', where clusters of separate languages have formed without long-term geographic or communicative isolation. In biology, 'geographic isolation is a purely extrinsic and completely reversible factor which does not by itself lead to the formation of species. Its role is simply to permit the undisturbed genetic reconstruction of populations that is the prerequisite for the building up of [reproductive] isolating mechanisms' (Mayr 1963:554). This principle would seem to hold true for language genesis as well. If two speech communities of the same language become geographically separated and thus communicatively isolated for a period, but are reunited before total divergence sets in, then presumably whatever differences have set in during the interim will sooner or later be reconciled and the language will not split. Thus we see a three-step process being the dominant mode of both speciation and language genesis: (I) migration or separation that results in geographical isolation for a long period, so that there develops (2) genetic/speech adaptations that result in reproductive/communicative isolation from the parent population, ultimately producing (3) selfrecognition (genetic or linguistic) by the group as a new species or language. Fourth, and most important, speciation and language genesis occupy similar positions of importance with respect to the creation and continuation of diversity. Speciation is potentially a process of evolutionary rejuvenation, an escape from too rigid a system of genetic homeostasis. ... The importance of.spe-, elation is that it invites evolutionary experimentation. It creates new units of evolution, particularly those that are important for potential macroevolutton. Speciation is a progressive, not a retrogressive, process' (Mayr 1963:555). So language genesis is with respect to cultural diversity. Culture requires continuity if it is to have meaning, and languages are the vehicle of continuation. Languages are the building blocks of cultural diversity, arguably the fundamental 'raw material
' of human thought and creativity. If their numbers are reduced dramatically, then the raw material for human creative evolution is diminished, eventually making the world's cultures increasingly monolithic, with the range of cultural variety severely circumscribed. Unfortunately, the continuation of intensive speciation and language genesis is threatened by the elimination of the conditions that historically made reproductive and communicative isolation possible. In fact, one could plausibly argue that
96 SOUTHWEST JOURNAL OF LINGUISTICS, VOLUME 15, NUMBERS 1 & 2 (1996) the conditions which allowed for the creation of the great number of species and languages we know today have already largely disappeared. In terms of species, we now have the unprecedented and pervasive invasion by humans into the remotest corners of the Earth. In many places the result has been habitat conversion on a massive scale and thus the current biodiversity crisis. In terms of languages, the dawning era of global telecommunications means that dominating languages can now be heard daily in the remotest corners of the Earth. Geographic isolation no longer necessarily produces communicative isolation. 3. SIMILARITIES IN THE DISTRIBUTION OF SPECIES AND LANGUAGES. Species and languages are not just comparable on an abstract, conceptual level. There is also a striking pattern of congruity in the geographical distribution of the two. For instance, many countries with high numbers of endemic species also have many endemic languages. ENDEMIC SPECIES are those found in restricted locales and nowhere else (WRI 1988:95). In discussions of global biodiversity, endemicity usually is pegged to individual countries (cf. WCMC 1992:137, 155). Endemic species are important because they represent unique adaptations to environmental conditions in a relatively small area. For this same reason they are vulnerable to predation or competition by less-specialized invasive exotic species. The concept can be readily extended to languages, so that ENDEMIC LANGUAGES are those restricted to a single country.' Like their species counterparts, endemic languages hold a high percentage of the unique traits in human language. This is borne out by analysis of the twelfth edition of Ethnologue(Grimes 1992), which reveals that just over 83% of the world's languages (5,635 out of 6,760) are endemics. (A detailed table listing them by country may be found in Harmon 1995.) They are also vulnerable to 'predation and competition' from dominating and invasive languages that are larger or more favored. 01 the 25 countries with the greatest number of endemic higher vertebrate species (i.e. mammals, reptiles, and amphibians), 16 are also among the top 25 in endemic languages-a concurrence of 64% (Table 1). It seems highly unlikely that this is mere coincidence. There are several geographical and environmental factors that explain why so many countries appear on both lists: (I) Extensive countries with highly varied terrain, climate, and ecosystems tend to have high numbers of endemic species simply because of their size and biophysical diversity. These same factors, operating at the lower population levels that prevailed before European expansion, also fostered communicative isolation among small speech communities, thus allowing many small autonomous languages to evolve. Examples from Table 1 include Mexico, USA, Brazil, India, and China. ' This is a strict definition, and it has its problems; for more, see Harmon 1995.
LOSING SPECIES, LOSING LANGUAGES
(2) Island countries tend to have high numbers (and often a high density) of endemics because their physical isolation from continental land masses has allowed more locally adapted species to develop. Islands that in addition have broken terrain or some other significant physical barriers to easy internal movement of peoples also will
TABLE 1. Endemism in language and higher vertebrates: Comparison of the top 25 countries
ENDEMIC LANGUAGES (n) 1.Papua New Guinea
(847) 2.Indonesia (655) 3.Nigeria (376) 4.India (309) 5.Australia (261) 6.Mexico (230) 7.Cameroon (201) 8.Brazil (185) 9.Zaire (158) 10.Philippines (153) 11.USA (143) 12.Vanuatu (105) 13.Tanzania (101) 14.Sudan (97) 15.Malaysia (92) 16.Ethiopia (90) 17.China (77) 18.Peru (75) 19.Chad (74) 20.Russia (71) 21.Solomon Islands(69) 22.Nepal (68) 23.Colombia (55) 24.Cote d'Ivoire (51) 25.Canada (47)
ENDEMIC HIGHER VERTEBRATE SPECIES (II) Australia (1,346) Mexico (761) Brazil (725) Indonesia (673) Madagascar (537) Philippines (437) India (373) Peru (332) Colombia (330) Ecuador (294) USA (284) China (256) Papua New Guinea(203) Venezuela (186) Argentina (168) Cuba (152) South Africa
(146) Zaire (134) Sri Lanka (126) New Zealand (120) Tanzania (113) Japan (112) Cameroon (105) Solomon Islands (101) Ethiopia (88) Somalia (88)
Notes: Countries appearing on both lists are in bold. Figures for Ethiopia include Eritrea. Higher vertebrates include mammals, birds, reptiles, and amphibians: reptiles not included for U.S., China, and Papua New Guinea, because the numbar of endemic species is not reported in the source table. Sources: Endemic languages figures derived from Grimes 1992. Species figures from WCMC 1992:139-141.
98 SOUTHWEST JOURNAL OF LINGUISTICS, VOLUME 15, NUMBERS 1 & 2 (1996) tend to have more endemic languages than similar continental countries. Examples from Table 1 include Papua New Guinea, Philippines, and Solomon Islands. (3) Tropical countries tend to have more species than those with predominantly boreal, temperate, Mediterranean, or austral climates (WCMC 1992:43). Tropical countries also tend to have more endemic languages than the others because the warm climate and ample rainfall, coupled with the richness of the natural resources at hand, was historically favorable to the small hunter--gatherer societies that flourished before the advent of concentrated agriculture. With these small cultural groups came many endemic languages. Examples from Table 1 include Cameroon, Zaire, and Tanzania. Where all three of these conditions coincide within a country, there comes the possibility of extremely high endemic richness in both species and languages. Indonesia, which ranks fourth on the vertebrate species and second on the language list, is the cardinal example of this. Australia, which ranks in the top five of both lists, combines some of the factors. Given that they make up such a high proportion of the overall total, the distribution of endemic languages may be taken to stand for the distribution of languages as a whole. If that is so, one might expect the distribution of endemic languages to match up well with that of species as a whole, both endemics and non-endemics. There is evidence that this is exactly the case. Nearly every region rich in species is also rich in endemic languages. Central Africa, Amazonia, southeastern Asia, Australia, and China conform to the pattern. The same is true on a national basis: the list of countries with the most endemic languages corresponds very well with a list of the TABLE 2.`Megadiversity' countries (listed alphabetically): Concurrence with endemic languages rank (in parentheses) Australia (5) Brazil (8) China (17) Colombia (23) Ecuador (--) India (4) Indonesia (2) Madagascar (--) Malaysia (15) Mexico (6) Peru (18) Zaire (9) Notes: The 12 `megadiversity countries' were identified on the basis of species lists for vertebrates, swallowtail butterflies, and higher plants. Source: McNeely et al. 1990.
LOSING SPECIES, LOSING LANGUAGES
`megadiversity countries' identified as having a large fraction of the world's species diversity in selected groups of animals and plants (McNeely et al. 1990). Table 2 shows that 10 of the 12 megadiversity countries (83%) are also among the top 25 in endemic languages. Finally, the pattern carries over to a comparison of endemic languages with plant species, where the concurrence between the top-25 lists is 16 of 25, or 64% (Table 3).
TABLE 3. Endemic languages vs. flowering plant species: Comparison of the top 25 countries
ENDEMIC LANGUAGES (11.) 1.Papua New Guinea(847) 2.Indonesia (655) 3.Nigeria (376) 4.India (309) 5.Australia (261) 6.Mexico (230) 7.Cameroon (201) 8.Brazil (185) 9.Zaire (158) 10.Philippines (153) 11.USA (143) 12.Vanuatu (105) 13.Tanzania (101) 14.Sudan (97) 15 Malaysia (92) 16.Ethiopia (90) 17.China (77) 18.Peru (75) 19.Chad (74) /0.Russia (71) ThSolomon Islands (69) 72.Nepal (68) /3.Colombia (55) 74.C6te d'Ivoire (51) 75.Canada (47)
FLOWERING PLANT SPECIES (n) Brazil (55,000) Colombia (35,000) China (30,000) Mexico (20,000-30,000) Venezuela (15,000-25,000) USSR (former)(22,000) Indonesia (20,000) Ecuador (16,500-20,000) USA (18,956) Bolivia (15,000-18,000) Australia (15,000) India (15,000) Peru (13,000) Malaysia (12,000) Thailand (12,000) Costa Rica (10,000-12,000) Zaire (11,000) Papua New Guinea (10,000) Tanzania (10,000) Argentina (9,000) Madagascar (8,000-10,000) Panama (9,000) Turkey (8,472) Cameroon (8,000) Philippines (8,000) Guatemala (83)00)
Notes: Countries appearing on both lists are in bold. Figutes fin- Ethiopia include Eritrea. Flowering plant species include both endemics and non-endemics. Sources: Endemic languages figures derived from Grimes 1992. Species figures from WCMC 1992:80-83.
100 SOUTHWEST JOURNAL OF LINGUISTICS, VOLUME 15, NUMBERS 1 & 2 (1996) The significance of endemicity in language is that it is an artifact of the historical condition of geographical and communicative isolation which made possible many of the languages we have today. Thanks to telecommunications, even in the remaining remote areas of non-industrialized countries this condition no longer exists. Therefore, endemicity in language--and, by extension, abundance and diversity in language, since endemics make up such a large percentage of the whole--can be likened to relicts of a lost world. Endemicity calls to mind another intriguing possibility: that numerous small cultural groups have co-evolved with the locally adapted animals and plants around them. While this possibility must remain quite speculative at this point, the intimate knowledge indigenous peoples often have of their immediate environment, and their traditional dependence on it, beckons us toward further consideration of co-evolution. WhateVer the case, traditional environmental knowledge is increasingly being seen as valid scientific information by biologists (see, for example, Wilson 1992:4244).2 Unfortunately, as more and more native groups are set adrift in the flood tide of globalized pop culture
, this knowledge runs the risk of being lost. Reports such as those chronicling the decay of the biological lexicon of the O'odham in Arizona (Hill 1995; see also Nabhan & St. Antoine 1993) are therefore distressing from both a biological and linguistic perspective. 4. SPECIES EXTINCTION AND LANGUAGE Loss. Many linguists have remarked On the similarities between the critical situation facing the world's biodiversity and its linguistic diversity (e.g. Zepeda & Hill 1991:135; Hale 1992; Krauss 1992). Wurm has very ably sketched them, and I can do no better than to quote him at some length: There are close parallels in the circumstances surrounding the decimation and eventual extinction of animal or plant species, and in those of languages. ... Extinction through violence and catastrophe is readily comparable in both instances, but the same is also the case with changes in ecology: an animal or plant species loses its viability and ability to survive through the drastic reduction or alteration of its habitat and/or the introduction of other animal or plant species which in some important respects are more powerful and with which the species concerned is unable to compete successfully. The introduction of predatory animals such as dogs and cats into areas in which the local fauna has no natural defences against them is a well-known example of the latter, and the replacement of an untodched wilderness by an agricultural area an example of the former. Both can lead to a reduction of numbers of the original animal or plant population to below a minimal level needed 'In a similar vein, linguists have used species terminology as a tool to help reconstruct both the migrations of speech communities and the structure of proto-languages (Hinton 1994:87-92).
LOSING SPECIES, LOSING LANGUAGES
for its survival and reproduction. All this can be readily translated into comparable circumstances surrounding languages: changes in environment would mean that the cultural and social settings in which a given language had been functioning, usually for a very long time, have been replaced by new and quite different ones as a result of irresistible culture contact and clash, with the traditional language unsuited for readily functioning as a vehicle of expression of the new culture. The newly introduced dangerous animal and plant species mentioned above can be compared with negative and destructive attitudes towards this traditional language by the carriers of the newly introduced culture and speakers of the language serving as its means of expression (1991:2-3). Let me expand on some of the parallels Wurm has drawn. The analogy between the destruction of natural habitat (for species) and the traditional social setting (for languages) is right on the mark. Both species and languages have evolved over hundreds or thousands of years to adapt to very specific contexts. If those contexts undergo unprecedented rapid change--as the world's environment and culture are now doing--many species and languages will likely lack the resiliency to adapt to the new conditions. In biology, island-dwelling species are hallmarks of such highly specialized, highly vulnerable life forms; and, sure enough, exactly 75% of all recorded animal extinctions occurring since 1600 have been of island species (WCMC 1992:199). Presumably, languages small in size and extent (and thus perforce among the most vulnerable to extinction) are so precisely because they have historically adapted to local conditions only. Conversely, certain species and languages show a great capacity to invade the habitat and social settings of others. A special kind of exotic species, the organisms involved in epidemic disease, is responsible for the loss of untold numbers of native languages whose speakers were wiped out after European contact. This was (and, in Amazonia, still is) an especially important factor in the Americas (see Adelaar 1991:45 for South America, Garza Cuaron & Lastra 1991:105 for Mexico, Zepeda & Hill 1991:136 for the U.S., and Kinkade 1991:157 for Canada). There is also a direct connection between a very small class of favored nonnative species--namely, the domesticated plants and animals used in agriculture-- and the death or serious decline of certain languages. Several hunter-gatherer groups in East Africa have given up their languages in favor of 'cattle languages' as part of a general biocultural deference to the higher-prestige occupations of pastoralism and farming (Dimmendaal 1989:16-21; 13renzinger et al. 1991:31, 39). Certainly the infamous Highland Clearances, which were tied directly to the desire to increase sheep holdings, helped put Scots Gaelic into the precarious position it is in today (cf. Hamp 1989:208). In addition to agriculture, recent history has seen the emergence of a global network of interconnected industries based on the extraction of natural-resource commodities (e.g. fishing, logging, mining).
102 SOUTHWEST JOURNAL OF LINGUISTICS, VOLUME 15, NUMBERS I & 2 (1996) Together, these are extremely potent homogenizing forces working to the detriment of biological and cultural diversity. The actions which fragment or destroy wildlife habitat also serve to homogenize cultures and languages: converting wildlands to pasture or cropland (which is in part driven by the burgeoning global export market for agricultural goods); building roads, rail lines, and air strips in remote areas; developing logging concessions, mines, and other industries in tribal areas, and so on. I might also briefly mention an issue that will loom larger and larger in the next century: urbanization. One of the defining demographic trends of our time is the increasing concentration of the world's population in cities. The most dramatic shifts to the cities are now taking place in the less-developed countries, where many endemic species and languages are concentrated. 'In terms of human psychology, there are profound consequences implicit in switching from a world in which most people grew up in rural areas--close to the land, so to speak--to one in which most grow up in cities' (Harmon & Brechin 1994:107). Might succeeding generations of city-dwellers become so detached from the natural environment that their interest in preserving it will eventually wane (cf. Lusigi 1988:44)? And could the same apply to urban migrants cut off from their ancestral communities and cultures? Among small-language communities in Africa where 'migration to the city has become a common pattern', it has been said that 'language death is a likely consequence' (Brenzinger et al. 1991:32). Some urban migrants may simply reject their former language, as have Nubian speakers in Cairo (Rouchdy 1989:96), or only the most highly motivated among them will try to keep it up, as has been reported with Irish and Scottish Gaelic (Watson 1989:45-46). 5. ASSESSING BIOLOGICAL AND LINGUISTIC DIVERSITY. Biological diversity (or 'biodiversity' for short) is the range of variation exhibited by life forms. The term is 'commonly used to describe the number, variety and variability of living organisms' and is essentially a synonym for 'life on Earth' (WCMC 1992:xiii). There are three fundamental, hierarchically related levels of biological organization at which biodiversity is measured: the genetic, species, and ecosystem levels. From an evolutionary perspective
, the genetic level is probably the most important because evolutionary change ultimately is reflected in gene frequencies (cf. Futuyma 1989:573). From a conservation management standpoint, the ecosystem level is arguably the most important because it also includes abiotic (non-living) components or natural communities and thus is the most holistic measurement. In practice, however, biodiversity is usually measured on the species level. Moreover, 'discussion of global biodiversity is typically presented in terms of global numbers of species in different taxonomic groups' (WCMC 1992:xiii), with the result that the term 'biodiversity' has become equated with 'species richness
'. This is a rather loose usage, since the sheer number of species extant is only an approximation of the range of taxonomic variation among them, though it is a good proxy measure nonetheless.
LOSING SPECIES, LOSING LANGUAGES
Linguistic diversity is the range of variation exhibited by human languages. It too can be measured on three fundamental, hierarchically related levels: the structural level, the language level, and the genetic (or lineage) level. Structural diversity has to do with the amount of disparity exhibited by the structural features and types within a language or population of languages (Nichols 1992:237). Genetic diversity refers to the 'number of discrete lineages and the extent to which individual lineages have branched out. The more branches or lineages, the greater the diversity' (Nichols 1992:232). Diversity at the language level most properly refers to the range of variation between individual languages across lineages, but the sheer number of distinct languages is a good approximation of linguistic diversity. Table 4 shows how the biological and linguistic levels correspond.
TABLE 4. Biological and linguistic levels for assessing diversity
BIOLOGICAL LINGUISTIC Genetic Structural Species Language
SIMILARITIES Scale of resolution: fine Potentially most accurate measurement of diversity Highly technical concepts;difficult for laypersons to grasp Scale of resolution: intermediate Richness used as proxy of diversity Species/languages less well defined than popularly thought Intuitive concepts; relatively easy for laypersons to grasp Most widely used measures of diversity
Ecosystem Genetic (Lineages)
Scale of resolution: broad Lack of consensus on how to define ecosystems/lineages; some classifications are thus problematical (see WRI 1990:123 and Ruh1ell 1991) Potentially more accurate measure of diversity than species/language level
I have already discussed the factors that promote endemism in species and languages: (1) extensive, varied terrain; (2) island geography; and (3) tropical climate. Nichols identifies essentially the same factors as promoting high genetic diversity among languages: High genetic diversity is evidently favored by coastline, tropical to subtropical climate, and (at least in some cases, such as the Caucasus and Himalayas) mountains. Low density is favored by high latitudes, drier and/ or more seasonal continental interiors, and also by the presence of large-scale economies and/or societies such as empires whose languages spread with their political/econoinic systems. ... Genetic diversity is notably high around the Pacific rim, where a number of these t'actors--coastline, in i Id climate, ab-
104 SOUTHWEST JOURNAL OF LINGUISTICS, VOLUME 15, NUMBERS 1 & 2 (1996) sence of empires (up to colonial times)--continue to make it possible for a relatively small community to survive autonomously (1992:234). Nichols includes the Pacific Rim among a number of 'residual zones'--'small regions of high genetic and structural density which may be thought of as capturing world linguistic diversity in microcosm' (1992:250-251). Since many of the same factors promote endemism and genetic diversity in language, we might expect that several residual zones to coincide with areas of high endemism. And they do: Australia, Papua New Guinea, Nepal, Canada and the U.S. (many of the endemic languages are situated in a belt running from California to British Columbia), and Ethiopia stand as examples. There are differences, however: the Balkans, Caucasus, and Pacific Coast of Northeast Asia do not show up on the high-endemism list. This serves as a potent reminder that no single measure of anything so complicated as linguistic or biological diversity is diagnostic. Other differences of opinion form part of a larger debate on where humans originated and how they colonized the world. The attempts of the geneticist Luigi Luca Cavalli-Sforza and various collaborators to solve this problem have put the complex relationship of biology and linguistics squarely in the middle of the arena of scholarly debate. Cavalli-Sforza's theory places the origination in Africa. He uses data on the distribution of genes,,corroborated by the distribution of languages, to reconstruct the subsequent migrations (Cavalli-Sforza 1991). A very different, much more strictly linguistic, approach to the origination-migration question is taken by Nichols (1992), whose conclusions differ considerably from CavalliSforza's. The upshot for linguistics is that she sees colonized areas of the Pacific Rim and the New World as the centers,t)f highest diversity (1992:250-53), while Ruhlen, supporting Cavalli-Sforza, places the center in Africa (1994:163-64, 108-9).3 If we ever arrive at a generally accepted solution (for it remains to be seen whose theory will carry the day), then presumably much of the mystery of human evolution will have been cleared away. In any event, the answer to the problem is obviously relevant to explaining present-day patierns of linguistic diversity. This would be true even if Cavalli-Sforza had made no overt reference to language or linguistics, but rather had confined himself to a comparison of genetic data with the distribution of ethnic groups or sonic other measure of cultural diversity. As it stands, his findings' support of Joseph Greenberg's controversial classification of New World language families, have brought him into conflict with many linguists. 'Others, not involved in the migration-origination debate, have still other ideas about where the center of linguistic diversity lies. Diamond points to New Guinea because the island holds about one-fifth of the world's languages (as reckoned by Ruhlen), representing 'dozens of independent languages or language stocks, with no proven relationship to each other nor to any other language stock in the world' (Diamond 1993:252-53, drawing upon Foley 1986). And Adelaar speaks in passing of South America's 'unsurpassed genetic variety', considering the relatively low number of languages there (1991:45).
LOSING SI'ECIES, LOSING LANGUAGES
TABLE 5. Some analogous concepts in biology and linguistics
BIOLOGY Subspecies Species Family Order Phylum Capacity to interbreed or unique combinations of character states Monotypic genus Invasive alien species (exotic) species Lack of reproductive capacity Circular overlaps (Mayr 1963:507-12) Clad istics Refugia
LINGUISTICS Dialect Language Family* Stock* Phylum* Mutual intelligibility
ANALOGY Variation within basic units of diversity Basic unit of diversity (cohesion) Group of related species (languages) Group of related families High-level taxon Simplified definition ofspecies/language
Language isolate High importance in diversity Dominating Can displace native species/languages language Moribundity Loss of capacity to pass genes/language to next generation; the 'living dead' Dialect chains Continuousgrading of changes producing distinctiveness at end of circle/chain Language family Use of tree diagrams to illustrate relationclassification among species/languages over long periods of time Residual zones Areawith accretions of diverse species/ 1,;litguages, including `strandings' (e.g. Hungarian); 'residual zones arelinguis- 'tic refugia where pre-spread stocks regularly survive' (Nichols 1992:237); cf. Pleistocene refugia
* These terms may be used ambiguously with refornce to language (Ridden 1991:21).
6. CONCLUSION: THE CASE FOR DIVERSITY. I hope to have shown in this paper that the comparison between species'and languages is not simply metaphorical, but instead has a basis in real-life shared processes that are reducing their numbers, in actual data on geographical distribution, and in a set of analogous concepts (outlined in Table 5) that have probative value in explaining diversity. After all is said and done, we Should never lose sight of the fact that diversity is an organizing principle. To assess diversity is to make judgments about How mucti species and languages. differ from each other, not merely to determine that they are different, and this iniplies a system to classify the differences. If making such distinctions between species and languages is important, it is important first of all because it helps us see the world in a more accurate light. 'Numbers alone
106 SOUTHWEST JOURNAL OF LINGUISTICS, VOLUME 15, NUMBERS 1 & 2 (1996) do not make science; it is relations between numbers that are needed', wrote the biologist Robert MacArthur. 'Applying a formula and calculating a "species diversity" from a census does not reveal very much; only by relating this diversity to something else-something about the environment perhaps-does it become science. Hence there is no intrinsic virtue in any particular diversity measure except insofar as it leads to clear relations' (1972:197). It is this illuminative power of diversity which is paramount. To grasp the magnitude and variety of biological and cultural wealth around us is to gain abetter perspective on our relationship to the rest of life. To be concerned about their loss is to better understand our responsibilities to other species and to the rest of our own. For biologists and linguists working to stop extinctions, diversity is a way to value the world. If, in turn, we are to value diversity, we must be able to grasp it intuitively. Species and languages, for all their faults and ambiguities, are the concepts that currently work best to measure diversity. The ultimate argument for preserving them is that they make the Earth a better place to be. In the end, the case for diversity must appeal to this value. 'We do not understand ourselves yet and descend farther from heaven's air if we forget how much the natural world means to us', writes Wilson. 'Signals abound that the loss of life's diversity endangers not just the body but the spirit.' We must strive, then, for an 'enduring environmental ethic' which aims 'to preserve not only the health and freedom of our species, but access to the world in which the human spirit was born' (Wilson 1992:351). As for cultural diversity, Whitehead (1941) put it simply and directly: 'A diversification among human communities is essential for the provision of the incentive and material for the Odyssey of the human spirit. Other nations of different habits are not enemies: they are godsends.' REFERENCES A DELAAR, WILLEm RH. 1991. The endangered languages problem: South America. In Robins & Uhlenbeck, 45-91. BRENZINGER, MATIHIAS, BERND HEINLE, and GABRIELE SOMMER. 1991. Language death in Africa. In Robins & Uhlenbeck, 19-44. CAvALLI-SFoRzA, Lui(n LUCA. 1991. Genes, peoples and languages. Scientific American 265.104-10. COLLINGE, N. E. 1990a. An encyclopaedia of linguistics. London and New York: Routledge. COLLINGE, N. E. 1990b. Language as it evolves: Tracing its forms and families. In CoHinge 1990a, 876-916. CRACRAFF, JOEL. 1989. Speciation and its ontology: The empirical consequences of alternative species concepts for understanding patterns and processes of differentiation. In Otte & Endler, 28-59.
LOSING SPECIES, LOSING LANGUAGES
DIAMOND, JARED. 1993. New Guineans and their natural world. In Kellert & Wilson, 251-271. DIMMENDAAL, GERRIT J. 1989. On language death in Eastern Africa. In Dorian, 13-31. DORIAN, NANCY D. 1989. Investigating obsolescence: Studies in language contraction and death. Cambridge, U.K.: Cambridge University Press. DURRELL, MARTIN. 1990. Language as geography. In Collinge 1990a, 917-55. FOLEY, W. A. 1986. The Papuan languages of New Guinea. Cambridge, U.K.: Cambridge University Press. (Cited in Diamond 1993.) FUTUYMA, DOUGLAS J. 1989. Macroevolutionary consequences of speciation: In- ferences from phytophagous insects. In Otte & Endler, 557-78. GARZA CUARON, BEATRIZ, and YOLANDA LASTRA. 1991. Endangered languages in Mexico. In Robins & Uhlenbeck, 93-134. GRIMES, BARBARA F. (ed.) 1992. Ethnologue: Languages of the world. 12th edi- tion. Dallas: Summer Institute
of Linguistics. HALE, KEN. 1992. On endangered languages and the safeguarding of diversity. Language 68.1-3. HAMP, ERIC P. 1989. On signs of health and death. In Dorian, 197-210. HARmoN, DAVID. 1995. The status of the world's languages as reported in Ethnologue. Southwest Journal of Linguistics 14.1-33. -, and STEVEN R. BRECIIIN. 1994. The future of protected areas in a crowded world. The George Wright Forum 11:3.97-116. HILL, JANE H. 1995. Language decay: The loss of structural differentiation ill obsolescent languages. Paper presented at the annual meeting of the American Academy
for the Advancement of Science, Atlanta, Georgia. HINTON, LEANNE. 1994. Flutes of fire: Essays on California Indian languages. Berkeley: Heyday. JESPERSEN, arm. 1922. Language: Its nature, development and origin. New York: Henry Holt. KELLERT, STEPHEN R. and EDWARD O. WILSON. 1993. The biophilia hypothesis. Washington, DC
: Island Press. KINKADE, M. DALE. 1991. The decline of native languages in Canada. In Robins & Uhlenbeck, 157-76. KRAUSS, MICIIAEL. 1992. The world's languages in crisis. Language 68.4-10. LAW, ViviEN. 1990. Language and its students: The history of linguistics. In CoHinge 1990a, 784-842. LUSIGI, WALTER J. 1988. The new resources manager. For the conservation or Earth, ed. by Vance Martin, 42-52. Golden, CO: Fulcrum. MACARTHUR, ROBERT H. 1972. Geographical ecology: Patterns in the distribution of species. New York: Harper & Row. MAYR, ERNsT. 1963. animal species
and evolution. Cambridge, MA: The Belknap Press of Harvard University Press.
108 SOUTHWEST JOURNAL OF LINGUISTICS, VOLUME 15, NUMBERS 1 & 2 (1996) MCNEELY, JEFFREY A., KENTON R. MILLER, WALTER V. REID, RUSSELL A. MITIERMEIER, and TimoTHY B. WERNER. 1990. Conserving the world's biological diversity. Gland, Switzerland, and Washington, DC: IUCN, WRI, Conservation International, WWF-US, and the World Bank
. NABHAN, GARY PAUL, and SARA ST. ANTOINE. 1993. The loss of floral and faunal story: The extinction of experience. In Kellert & Wilson, 229-50. NiChoLs, JOHANNA. 1992. Linguistic diversity in space and time. Chicago: University of Chicago Press. Orrri, DANIEL, and JOHN A. ENDLER (eds.) 1989. Speciation and its consequences. Sunderland, MA: Sinauer. PEDERSEN, HOLGER. 1962. The discovery of language: Linguistic science in the 19th century. Bloomington: Indiana University Press. Original English edition 1931. ROBINS, ROBERT H., and EUGENIUS UHLENBECK. 1991. Endangered languages. Oxford and New York: Berg. RuHLEN, MERRITT. 1991. A guide to the world's languages, Volume 1: Classification. Reprint, with a postscript on recent developments, Stanford, CA: Stanford University Press. Original edition, Stanford, CA: Stanford University Press, 1987. -. 1994. The origin of language: Tracing the evolution of the mother tongue. New York: John Wiley & Sons. TEMPLETON, ALAN R. 1989. The meaning of species and speciation: A genetic perspective. In Otte & Endler, 3-27. -. 1991. Genetics and conservation biology. Species conservation: A populationbiological approach, 15-29. Basel: Birkhauser. (Cited in WCMC 1992.) WATSON, SEOSAMIL 1989. Scottish and Irish Gaelic: The giant's bed-fellows. In Dorian, 41-59. WCMC [World Conservation Monitoring Centre]. 1992. Global biodiversity: Status of the Earth's living resources. London: Chapman & Hall. WHITEHEAD, ALFRED NORTH. 1941. Science and the modern world: Lowell Lectures, 1925. Reprint. New York: Macmillan. WILSON, EDWARD 0. 1992. The diversity of life. Cambridge, MA: Belknap Press of Harvard University Press. WRI I World Resources Institute and International Institute for Environment and Development
]. 1988. World resources 1988-89. New York: Basic Books. WRI World Resources Institute]. 1990. World resources 1990-91. New York: Oxford University Press. WI EM, STEPHEN. 1991. Language death and disappearance: Causes and circumstances. In Robins & Uhlenbeck, 1-18. ZEPEDA, OFELIA, and JANE H. Hu,. 1991. The condition of Native American languages in the United States. In Robins & Uhlenbeck, 135-55.
s in Top 25 for
s in Top 25 for vertebrates
s in Top 25 for both
Cuba Venezuela Brazil
Cote d'Ivoire Cameroon Dem. Rep. of Congo
996, based on data from Groombridge 1992 (139-141, for species) and Grimes 1992 (passim, for languages).