sign in sign up
<<
Home , Comparative linguistics, Joseph Greenberg, Macrofamily

S chrryer - Proto-Human 53

A Proto-Human Language: Fact or Fiction?

established will also be discussed with consideration of Proto-Human, and the The question of where and when idea that "all the spoken on anatomically modem humans emerged has earth (roughly 5,000...) are descendant of a long been a controversial issue in single ancestral language" (Ruhlen 1994b). anthropology. Evidence from many This paper will describe comparative different disciplines has been increasingly , and the many controversies useful in attempting to pinpoint the which are present in this type of study as origins of Homo sapiens in time and well as give examples of linguistic families place. As it is suggested that anatomical which have been compiled. It will also modems were the first humans to have look at genetic evidence, and the ability of the cognitive capacity for modem this to support the linguistic data. linguistic capabilities (Lieberman 1984), it Examples for and against the ability for may be plausible to use linguistic evidence to compile a in analyzing this biological evolution. proto-language that encompasses all This is because one of the main tenets of human languages will also be discussed in is that "languages an attempt to determine how far these spread as people move" (Barbujani and studies are viable into the past. Pilastro 1993). In fact, Dolukhanov writes in his examination of the archaeological record that: Comparative linguistics, sometimes The advent of language in the called historical linguistics, is often used in Upper Paleolithic was a powerful attempts to link languages that have factor that through the similar , , and intensification of information grammatical structures together. These transfer both in time (from one languages which are grouped together are generation to the next) and space called language families, and are said to be (between neighbouring groups) linked genetically. The languages which drastically accelerated social and are related are then combined into proto- cultural evolution. (1993). languages, or languages which are "reconstructed out of the evidence that is The purpose of this paper is to determine acquired by the careful comparison of the whether or not linguistics can truly be daughter languages [or lower level used to uncover the origins of anatomical languages]" (Haas 1978). These proto- modem humans, or if there is a time limit languages are only estimations of original at which linguistic comparisons can no languages, however, because it is longer be made. This timeframe within impossible to know exactly which which linguistic comparisons can be characteristics from each daughter S chrryer - Proto-Human Language 54 language were present in the original Another problem in the field of language. Proto-languages of spoken or comparative linguistics is the basic recorded languages are fairly accurate assumption that comparisons "account for although estimations which move further similarities which cannot be attributed to away from the actual data are more chance"(Hock and Joseph 1996), and are distorted, and are therefore less therefore the result of a common demonstrable. This is one of the major ancestor. The controversy stems from a problems with comparative linguistics of belief that the cognates which deep-timeframes, and is an essential fact comparative linguists use are not a to remember when attempting to compile product of related languages,but are due an original languagewhich may have been to the processes of diffusion, borrowing, ancestral to all languages on earth and onomatopoeia. These aspects do although advocates of this theory such as occur, but are relativelyeasy to identify as Greenberg and Ruhlen do not seem to chance rather than the result of genetic realize this. Another problem with relations while lexical similarities which historical linguistics is the lack of an are systematic and recurring are more absolute chronology. likely to be due to relations as are the Colin Renfrew writes that, "there presence of idiosyncrasies in languages seems nothing, however, in the pattern of (Hock and Joseph 1996). Borrowing is linguistic change which would allow for the use of words from one language into the establishment of any kind of absolute another, and "will generally be found in chronology" (1990). He suggests that language families that are geographically linguists take it for granted that languages adjacent...(though prehistoric adjacency change at a constant rate, and that might be hard to judge)"(Nettle 1999). It theories based on this premise such as stands to reason then that languages that are erroneous in nature. are geographically far apart will have less Glottochronology was a formula originally of a chance of having similarwords unless developed by in the they truly are related. Borrowing also 1960's as a means to "predict the tends to be limited to certain parts of of the derivation from a common vocabulary which are often technical in ancestor of a group of interrelated nature. In contrast similar words which languages reduced to sample wordlists" are part of any basic vocabulary are likely (Guy 1980). It has fallen under universal to be the result of genetic relationships criticism, however, partially as a result of (Hock and Joseph 1996). Onomatopoeic "too much theorizing and little expressions or sounds of nature also tend experimenting" (Guy 1980). This still to be remarkably similar in languages leaves comparative linguistics without a which are not related as are "nursery model for chronology although there have words" or words "which adults assign to been many attempts to find one. There is the early babbling of infants" (Hock and also the question of w~ languageswould Joseph 1996). diverge at a constant rate when there are The higher the number of numerous reasons why they would not languages which can be compared results such as geographical location and in the probability of fewer chance isolation, population densities, and the similarities. However, using comparisons state of relations within the speech of too many languages can also be community. problematic as is seen in the criticisms of 's methods of mass 5 chrryer - Proto-Human Language 55 comparisons to create exceptionally large has not only reached the furthest proto-languages from families of geographical regions from which it came, languages which are combined into lower but also because it "shows greater internal level proto-languages. The mass differentiation" (Greenberg, Turner, and comparison or multi-lateral comparison Zegura 1986). The second migration is method establishes genetic affinities of the arrival of people who speak Na-Dene, languages through comparison of a large and who populated mostly the North number of languages on a fairly limited West coast of North America, and are basis (Hock and Joseph 1996). This is therefore geographically less peripheral related to one last criticism of comparative than the third language group of Eskimo- linguistics which is that much of the work Aleut. The arrival of the Na-Dene has been completed by one highly language group appears to be dedicated individual, and for work to be approximately 5, 000 years B.P. (Krauss more widely accepted it is necessary for 1973). The third migration which is movement beyond "the great scholar" probably the most recent around 4,000 B. (NettIe 1999). This is partially due to the P. is the one which coincides with the biases which individual researchers may linguistic family of Eskimo-Aleut, and bring to their comparisons. People with whose populations live in the different knowledge may come up with Northernmost part of North America. different results even when using the same The languages and genes of this linguistic methodology (NettIe 1999). group have been shown to have relationships with Asian populations. EXAMPLES OF PROTO- Amerind is the most controversial LANGUAGE of the groupings, and this is partially due to the fact that it contains the highest Greenberg has used his multilateral amount of language families. As well it is comparison method in the midst of much thought to be of the greatest antiquity. criticism to suggest that the Americas are There is also the controversy over the fact compiled of three different linguistic that Greenberg uses oral languages which groups each of which coincides with a have not all been extensively researched to separate migration into the New World. complete his comparison. Arguments Since traditional linguists suggest that against his method include the idea of comparisons of languages cannot be made Pan-Americanism which is the belief that realistically past 7,000 years they eliminate there is a similarity of words in North the possibility of genetic affmity between American languages simply due to American languages and populations with similarities between Native cultures. As Asian languages and populations as this well, the general belief which most predates the arrival of people into the linguists hold is that there is simply too New World. Greenberg's comparisons much to be compared, and that negate this notion, and support the Greenberg can not thoroughly compare relationship between the two groups of all of the languages which are classified as people. The first migration which he Amerind. Critics often point out errors in suggests took place approximately 11,000- transcription and other minor details to 12,000 years ago, and coincides with the suggest that Amerind is a poor quality proto-language of Amerind. Amerind is comparison although few have the believed to be the oldest language evidence which would suggest the whole migration into North America because it work was not credible. In fact Greenberg Schrryer- Proto-Human lAnguage 56 summanzes his article In Difense if may be the way to eliminate these doubts Amerind by quoting Lamb who writes, from the minds of many scholars "The volume calls for careful study and especiallywhen these other sources might for follow-up research. It does not call provide concrete chronologies and for criticism based on incomplete observable changes throughout evolution understanding of his methods" (1987). of people which historical linguistics can Another proto-language which is not see. Raimo Antilla writes that on the same levelas Amerind is the proto- "collaboration between archaeologists and language of Nostratic which is a term that linguists is desirable because each side is somewhat ethnocentric meaning tends to apply the findings of the other "belonging to us/to our part of the world too simplistically"(1989). i.e. plus Northern Africa" (Hock and Joseph 1996). The validity of Nostratic, like Amerind, is also not accepted by the majority of historical For a long period of time there linguistics,but "if it is not proven, it is not has been the hope that "evidence from discredited either" (Nettle 1999). Again, it molecular genetics will cast more light is the lack of a concrete methodology upon population which may in which gives issue to the controversy over turn have a bearing on language history Nostratic, and similarly to Amerind, also" (Renfrew 1998). The use of genetic Nostratic's biggest critics are convinced evidence for linguistic comparisons is that the Proto-language contains too many discussed in this section. Genes and languages, and that the similarities languages have many similar traits in between them are just random. However, common when it comes to the way in unlike Amerind the languagesof Nostratic which they spread, and the ways in which are spread across many different regions, they are studied consequently. Both and because the origins of people in these utilize tree diagrams in an attempt to regions are not the result of a migration in relate the genes/languages of today back fixed time and place like the arrival of to an original source as well as show the people into the New World critics of "representation of inherited relationships" Nostratic claim that the combination of (McMahon, Lohr, and McMahon 1999). the languagefamilies is less accurate., The One fundamental difference which original compilation of Nostratic was also McMahon, Lohr, and McMahon note in completed by one scholar, Aharon their article "Family Trees and Favorite Dolgopolsky, and this is also cause for Daughters" is that biologistsuse trees that concern as many other linguists including are more based on quantitative data while Lyle Campbell find serious problems with linguists work more intuitively(1999). "nearly all of Dogopolsky's 124 Nostratic Cavalli-Sforza has been the main lexical sets" (1999). There are still many biologist who has done studies which link doubts about the soundness of Nostratic, these two separate phenomenon together. and unless it can be supported with more He suggests that the spread of genes and evidence linguists will continue to have languages are linked together, and that these doubts, and "any interpretations of studies of mtDNA enable patterns of non-linguistic based on it gene diffusion to be isolated to individual remain sheer speculation" (Campbell populations. Thus, if a gene is in one 1999). Use of non-linguistic sources to population, and moves to another, but not corroborate these linguistic comparisons to a third then those populations that Schrryer- Proto-Human Language 57 have the same gene must also have similar areas which link linguistic groups to languages, and consequently cultures. populations are also used in attempting to Cavalli-Sforza also notes that there are prove comparative linguistic studies. fewer linguistic trees than genetic ones, These include theories by Turner on but "it is usually true that the genetic dental evidence, and Zegura on genetic similarity between populations belonging differences such as blood types which to the same linguistic family is high" indicate that there were three distinct (1997).There are a few exceptions to this migrations into the Americas, and the rule including the populations of Lapps, populations of these migrations correlate Ethiopians, and Tibetans each of which with Greenberg's proposal of three speak a language to which they are not linguistic groups of Eskimo-Aleut, Na- biologicallylinked. These exceptions may Dene, and Amerind (Greenberg, Turner, be related to gene flow and language and Zegura 1986). Similarwork has been replacement, and while they "blur the done which connects Nostratic to genetic genetic and linguistic picture...they do not populations of Eurasia and North Africa; obscure it entirely" (Cavalli-Sforza1990). the areas from which the languages of Cavalli-Sforza also considers the place of Nostratic are taken (Barbujaniand Pilastro origin of the first populations of 1993). There have also been anatomical modems by examining how demonstrated genetic and linguistic closely related populations are to one population convergences in Australia another. Cavalli-Sforza indicates his (Ruhlen 1994b),and Southeast Asia (Baer believe that further collaboration between 1995). genetics and linguistics will be beneficial in examining the origins of modem FOR AND AGAINST PROTO- humans. There are similar problems HUMAN within each discipline as the rate at which genes and languages change is still The current view of most controversial, and the rate at which genes historical linguists is that there is a time- change is one of the main differences limit past which languages can not be between the continuity and replacement reconstructed. This limit has been models of human origins. suggested at different times which vary In respect to this genetic evidence drasticallyfrom 5,000 (Hock and Joseph Cavalli-Sforza indicates that all 1996), to 6,000 (Ruhlen 1994a; Ehret populations are related, and that Africa 1999), to 7,000 years (Greenberg 1996) and Asia are more closely linked than depending on the study. All this variation Europe (1997). As well, one of the between different scholars beliefs of when earliest branches in the evolution of linguistic comparisons can no longer be modem humans is the separation of done appears to indicateuncertainty in the Mricans and Non-Africans (1997). This field, and suggests that perhaps the time- suggests an African Origins Model, and limit is not so earlyas it was once thought would therefore suggest a replacement to be. Few linguistswould admit to this, model. Linguistic evidence also suggestsa however, who do not already believe in replacement model (Ruhlen 1994b)as it is the deep-time comparisons of languages with the arrival of anatomical modems leading to proto-languages such as that the cognitive capacity for languageis Nostratic and Amerind. Some linguists present suggesting monogenesis of all such as Ruhlen and Greenberg, as well as languages. Genetic studies of specific a few others, suggest taking comparisons Schrryer- Proto-Human Language 58 of language even further than this to (Hock and Joseph 1996). Greenberg and create what might be known as proto- Ruhlen fmd this in "Amerind, human where the "ultimate goal [is] a Eskimo-Aleut, and four language families comprehensive classification of what is of the Old World: Mro-Asiatic, Indo- very likely a single " European, Uralic, and Dravidian" (Hock (Greenberg 1987). The construction of and Joseph 1996);it appears in thirty-two this proto-language is situated in the belief languagesin all. Even with this suggested that: similarity in many languages numerous Human language linguists find problems with the came into being just once and of the cognate as they include such that all languages that now varying meanings as, "swallow, throat, exist (or ever have existed)are suck, chew, milk, breast, and neck" (Hock (or were) altered later forms and Joseph 1996). of this original language" Other indications that these global (Ruhlen 1994b). might be incorrect is seen in the article written by Trask which states Greenberg and Ruhlen have even that he can fmd fifteen similarities suggested 27 global etymologies which between Basque, a reputed language may in fact be evidence of proto-human, isolate, out of the 124 cognates described and Ruhlen suggests that these are only a in the proto-language of Nostratic (1999). fraction of what could actuallybe found if If a language which is known for being closer comparisons were made (Ruhlen isolated linguistically,and genetically (due 1994a). to the highest frequency of Rh-negative It is my belief, however, that gene of any population (Ruhlen 1994b)) eventually a point is reached where proto- can be seen to have similarities with languages come to the extent of the data higher levelproto-languages the legitimacy that the languages of present provide, and of this type of study for deep time-frames any other comparisons beyond this are is calledinto question. It is also important too far away from the original data to be to note that the Basque language survived correct. To summarize, the words of the the migration of Indo-European proto-language are so distorted that it is languages into its region (Ruhlen 1994b). impossible to tell whether the connection This suggests that Basque is the exception between cognates is a true one or not to the rule that languages spread with because we will never know if the proto- people, and this challengesthe assumption languages are even close to languages of language replacement in all territories which were spoken in the past. This except previously non-populated ones belief is supported with data in Hock and (Renfrew 1990). Joseph's (1996) book which suggest that Pessimistic views of this sort the global etymologies of Proto-human as which doubt whether there is enough suggested by Greenberg and Ruhlen are evidence to support proto-human "overly short". This is problematic in the languages have been around for a long sense that chance similarities are more time. As earlyas 1867 American historical likely to occur in short words than in linguist William Dwight Whitney wrote, longer ones (Hock and Joseph 1996; "Linguisticscience is not now, and cannot Moravcisk 1978). The one ever hope to be, in a condition to give an which is not too short is the proto-human authoritative opinion respecting the unity term maliq'a meaning throat or swallow or variety of our species". This is not to Schrryer- Proto-Human Language 59 say that linguistics, archaeology, and evidence to date on the use of linguistics genetics are not beneficial to each other to prove time and place of human origins: but the evidence which linguistics can If the lumpers (notably the provide is set within a limited time-frame. Russian school including Dolgopolsky ..., As Ruhlen writes "we know a good deal and the American school of Greenberg about what our ancestors looked like, and and Ruhlen) are correct, then the unity of how they lived, but their minds and human origins may be reflected in the languages remain shrouded in the past" evidence offered for linguistic (1994b). monogenesis ...But if the splitters are right, and if it is indeed the case that languages evolve so fast that no reconstructions of any kind could be possible beyond about This paper has discussed what 5,000 years ago ..., then, these broad comparative linguistics are, and the macro-families (e.g. Amerind, Austric, timeframe in which they can be applied, as Nostratic or Eurasiatic, Sino-Caucasian) well as why it is highly unlikely that use of would be entirely illusory, and so too in linguistic evidence can be traced back as consequence would any supposed far as the origins of modem humans. correration between them and the genetic This essay argues that linguistic evidence evidence for human phylogeny ... (1992). is useful, however, in showing that populations of today are related, and how I personally believe that a single origin of these changes in languages have occurred all human languages does exist, but trying as populations have spread across the to prove its existence is another matter as globe. The languages of Amerind are is attempting to describe it. The data especially useful to study as they show eventually gets too thin to postulate a how languages spread across a previously reliable version of a proto-language which uninhabited continent (according to encompasses all of the world's languages archaeological evidence), and at what rate. and consequently the language of the Although the true diversity of all original anatomically modem humans. languages spoken can not really ever be known approximate rate of change is an important key to determining how far back languages might be related even if Antilla,Raimo 1989. Historical and Comparative they cannot physically be connected. This Linguistics. Amsterdam and New York essay has also looked at the place of (NY):Benjamins. genetics in collaboration with linguistics, and how the combination of the two Baer,A. S. 1995. Human Genes and biculturalhistory in Southeast Asia. Asian pieces of data provide for a fuller Perspec!J:ves. 34(1):21-35. understanding of how anatomical modems spread at later time-periods than Barbujani,Guido and Andrea Pilastro 1993. 15,000 years ago. Whether or not Genetic Evidence on origin and dispersal Ruhlen's notion of a proto-human of human populations speaking languages of language is a reality or not is difficult if the Nostratic . Froc. Notl. not impossible to determine using the Acad. Sci. USA~ 90: 4670-73. evidence of the present. I quote Colin Renfrew's current appraisal of the Campbell,Lyle 1999. "Nostratic and linguisticpalaeontology in methodological Perspectives." In Nostratic: Examining a Schrryer- Proto-Human Language 60 Linguistic Macrofamify. Colin Renfrew Language History, Language Change, and and David Nettle, (ed.). pp. 179-230. Language Relationship: An Introduaion to Cambridge: McDonald Institute for Histon'cal and Comparative Linguistics. Berlin: Archaeological Research. Mouton de Gruyter.

Cavalli-Sforza, L. L. 1997. Genes, Peoples, Krauss, M. 1973. "Na Dene." In Cumnt and Languages. From the Colloquiwn: Trends in Linguistics. T. A. Sebeok, (ed.). pp. Genetics and On'gins 0/the Species. California: 903-78. Vol. 10. The Hague: Mouton. National Academy of Sciences. 94: 7719- 24. Lamb, Sydney M. 1987. Some proposals for linguistic taxonomy. Anthropological --- 1990. Comment. CumntAnthropology. 31: u'nguistics. 1(2): 33-49. 16-18. Lieberman, P. 1984. The Biolol!J and Evolution Dolukhanov, Paul, M. 1993. "The 0/Language. Cambridge: Harvard Pleistocene-Holocene Boundary: University Press. Environmental Processes and Social Adaptations." In From Kostenki to Clovis: McMahon, April, Marissa Lohr, and Robert U~ff~~~uand~~~~~Ad~~@~ McMahon 1999. "Family Trees and Olga Soffer and N. D. Praslov, (ed.). Favorite Daughters." In Nostratic: pp.189-196. New York: Plenwn Press. Examining a L:nguistic Macrofamify. Colin Renfrew and David Nettle, (ed.). pp.269- Ehret, Christopher 1999. "Nostratic or 85. Cambridge: The McDonald Institute Proto-human?" In Nostratic: Examining a for Archaeological Research. Linguistic Macrojami~. Colin Renfrew and David Nettle, (ed.). pp.93-112. Moravcisk, Edith 1978. "." Cambridge: The McDonald Institute for In: Universals o/Human Language. Joseph Archaeological Research. Greenberg, (ed.). pp.93-122. California: Press. Greenberg,Joseph 1996. In Defense of Amerind. Intern~onalJournal 0/Amen'can Nettle, David 1999. "Towards a future Linguistics. 62(2): 131-64. history of macro family research." In: Nostratic: Examt'nt'ng a Linguistic Macrojamify. --- 1987. Language in the Amen'cas. Stanford: Colin Renfrew and Daniel Nettle, (ed.). Stanford University Press. ppA03-19. Cambridge: The McDonald Institute for Archaeological Research. Greenberg, Joseph, Christy Turner, and Stephen Zegura 1986. The Settlement of Renfrew, Colin 1998. "Introduction: The the Americas: A Comparison of the Nostratic hypothesis, linguistic Linguistic, Dental, and Genetic Evidence. , and prehistoric studies." In: Cumnt Anthropology. 27(5):477-497. The Nostratic Mocrofamify and Linguistic Palaeontolol!J.A. Dolgopolsky, (ed.). pp.vii- Guy,]' B. M. 1980. Expen'mental xxii. Cambridge: The McDonald Institute Glottochronolol!J: Basic Methods and Results. for Archaeological Research. Canberra: Pacific Linguistics. --- 1992. Archaeology, Genetics, and Haas, Mary 1978. Language, CuIJure, and Linguistic Diversity. Man. 27:445-78. History. Stanford: Stanford University Press. --- 1990. "Archaeology and Linguistics: Some Preliminary Issues." In When Worlds Collide: Indo-Europeans and Pre-Indo- Europeans. Colin Renfrew, (ed.). pp. 15-24. Michigan: Karoma Publishers, Inc.

Ruhlen, Merritt 1994a. On the Origins of Languages: Studies in Linguistic Taxonomy. California: Stanford University Press.

--- 1994b. The On'gin of Language: Tracing the Evolution of the . New York: John Wiley & Sons, Inc.

Trask, R. L. 1999. "Why should a language have any relatives?" In: Nostratic: Examining a Linguistic Macrofami!J. Colin Renfrew and David Nettle, (ed.). pp.157-76. Cambridge: The McDonald Institute for Archaeological Research.

Whitney, William D. 1867. Language and the Stu4Y of Language. New York.