Looking for answers

We dealt in some way or other with questions that were in the present section we will look for answers—at least as far as this is possible and feasible on the basis of the narrow framework. To this end we will now deal with each of those questions in turn.

(a) Why did human language evolve, and what purpose did it serve? This question has occupied students of language evolution for centuries, and we do not attempt to summarize all the answers that have been volunteered, especially since none of them is really satisfactory; see Johansson (2005: 193–218) for a convenient summary (see also "Previous work"). In search for plausible reasons for why language evolved, a large variety of factors have been invoked, including economic activities such as cooperative hunting or new techniques of foraging (e.g. Bickerton 2005), cultural achievements such as tool-making (e.g. Wildgen 2004), new forms of social interaction such as the transition from grooming to gossiping (Dunbar 1996), human niche construction (Odling-Smee et al. 2003), or cognitive and neural factors, such as the growth and/or a modification of the human brain.

From the point of view of grammaticalization, the most promising hypothesis is one in terms of Darwinian adaptation and natural selection (Pinker and Bloom 1990); however, so far it has not been possible to develop a plausible hypothesis on what the nature of selection pressures was that led to adaptation. Bickerton (2005) observes that no other species has language and if any adaptation is unique to a species, the selection pressure that drove it must also be unique to that species. This may well be so, but presumably the situation characterizing language genesis was more complex. Findings on grammaticalization suggest that the rise of new functional categories is the result of a complex interaction of cognitive, pragmatic, and morphosyntactic variables. Very likely therefore there was not just one kind of selection pressure that was responsible for language genesis; rather, we suspect that there was a number of different forces that jointly contributed, or ‘‘conspired’’, to create human language.

But, discussion was limited to one particular issue, namely to what purpose early language may have served. The conclusion that we reached in "Functions of early language" was that communication must have been the primary motivation for creating early language, while cognition (‘‘inner speech’’, monologue, etc.) is likely to have served an important auxiliary function in structuring early linguistic communication.

(b) When and where did early language evolve? There is as yet no empirically sound answer to (b), which raises an issue that cannot be decided purely on the basis of linguistic evidence. Estimates on when humans began to talk range from between 2,000,000 and 9,000 years ago and— unfortunately—none of these estimates can really be falsified. Is the appearance of stone tools some 2.4 million years ago an indication that language was on the way in? Can the fact that our ancestors became anatomically modern humans and were able to comprehend and apply the concept of part–whole relationships to tool-making some 300,000 years ago (Wynn 1979; Lock 1988) offer clues on language origin, or does the explosion in culture and the size of population, the creation of art, and burying the dead roughly 40,000 to 60,000 years ago provide the proof required that there was language (see, e.g., Li and Hombert 2002)? Claims that core syntax must have been in place not later than 90,000 years ago (Bickerton 2005), or arose only some 50,000 years ago (Krantz 1980; Klein and Edgar 2002), are as good as any other of the many guesses that have been made on this issue. And whether there was already some form of rudimentary language (‘‘protolanguage’’) spoken by Homo erectus some 1.6 million years ago (Bickerton 1990) is a question that at the present stage of research is largely a matter of belief, that is, it is not testable in principle (see "The present approach"). And finally, even the questions of where the ultimate home area of our ancestors and of human language are to be located have so far not received an entirely satisfactory answer. There is good evidence that Africa was the cradle of the human species, and that human languages must also have originated in Africa; but even these assumptions are not entirely uncontroversial (Dennell and Roebroeks 2005).

Recent research on mitochondrial DNA lineage patterns, on behavioral innovations such as the emergence of hunting missiles, perforated shell beads, bone tools, and abstract geometrical designs that were found in southern African sites (Blombos Cave, Klasies River, and Diepkloof in particular) might be taken as providing converging evidence that the period between 85,000 and 50,000 BP was crucial for the rise of modern human languages (e.g. d’Errico et al. 2005; Mellars 2006; Henshilwood 2006). But it seems that there is still a long way until all this evidence can be turned into a viable hypothesis on the time and the location of early language evolution.

(c) Who were the creators of early language? With regard to this question, our discussion was restricted to one particular issue, namely to whether the creators of early language were young children, as has been claimed or implied in a number of works (e.g. Piattelli-Palmarini 1989). Our observations in "Who were the creators of early language?" suggest that it must have been adults and adolescents that created early language, while young children presumably played an important role in the second part of grammaticalization, making existing grammatical forms more regular, and being responsible for an increase of frequency of use, fluency, and fastness of production.

(d) Was its origin mono-genetic or poly-genetic, that is, do the modern languages derive from one ancestral language or from more than one? It has been argued that it makes no difference whether language began in one place or several (Bickerton 2005). We find this claim, which is based on specific theoretical assumptions, problematic. Observations on grammaticalization can be reconciled with both a mono- and a poly-genetic hypothesis. There are many examples showing that one and the same grammaticalization process has taken place independently in different parts of the world and in different historical periods (Heine and Kuteva 2002a). Since the conceptual and pragmatic principles underlying such processes are the same across languages, it is conceivable that grammatical structures as we know them from modern languages arose more than once in human evolution. But this is no more than a possibility; there simply are no appropriate data to answer (d).

(e) Were the forms and structures characterizing early language iconic or arbitrary? Another widely discussed issue is whether the lexicon and grammar at the early stages of language evolution were motivated in some way, for example by semiotic or pragmatic principles, or whether they were arbitrary (unmotivated) creations, showing no systematic correlations between form and meaning. Givo̒n (2002a: 4, 2005) assumes that language evolution led from more iconic, more indexical, less arbitrary to more arbitrary linguistic codings (see also Newmeyer 1991). One piece of argument can be seen in the rise of some restricted linguistic systems. Goldin-Meadow (2002: 348) discusses the results of a study of ten homesigning children, born to hearing parents who did not educate the children using an oral method. The gesture systems that these deaf children created consisted of pairings between gesture forms and meanings which, although having arbitrary aspects, were based on an iconic framework. And Givo´n (2002b: 29) finds evidence for this directionality, for example, in the fact that frequent use of well-coded signals inevitably leads to automation, speed-up, signal reduction and ritualization, and he draws attention to the development of writing systems, ‘‘whose time course in all five known centers where literacy is known to have arisen independently—China, India, Mesopotamia, Egypt, Maya—followed the same gradual move from early iconicity to later abstraction, arbitrariness and symbolization’’. That the coding of sequences of events in early language was largely iconic, in that sequences of events were presented in speech in the order they happened, would in fact seem to be a plausible assumption. Accordingly, with the emergence of morphology for clause combining, iconicity is likely to have lost in significance.

But it would seem that the role played by iconicity in early language requires much further research; whether the first lexical and functional forms to arise in human language were indexical, iconic or symbolic is an issue that must remain unresolved at the present state of research. Quite possibly, the use of non-iconic form–meaning units was not a human invention. As we observed, the alarm calls of vervet monkeys and other animal species are arbitrary forms (see, e.g., Cheney and Seyfarth 1990, 1992), and with appropriate training, a number of non-human animals have been found both to comprehend and produce what can be called elementary arbitrary symbols (see, e.g., Pepperberg 1999b: 43; Savage-Rumbaugh et al. 1980; Savage-Rumbaugh and Lewin 1994: 160).

A better understanding of this general issue depends to quite some extent on how the next question is to be answered.

(f) Did language originate as a vocal or a gestural system? Gestural and signed languages are generally characterized by a higher amount of indexical and iconic coding than spoken languages, and the present question falls, in principle, within the scope of our methodology. But our reconstructions were overwhelmingly restricted to speech, which is the modality that is most readily accessible to analysis and diachronic reconstruction; hence, we have little to say about the evolution of gestures and signing. Still, as we saw, grammaticalization theory has also been applied successfully to sign languages (e.g. Janzen 1995, 1998, 1999; Janzen and Shaffer 2002; Wilbur 1999; Shaffer 2000; Morford 2002; Sexton 1999; Pfau 2004; Pfau and Steinbach 2005a, 2005b). What surfaces from this research is that signed languages appear to develop in accordance with the same parameters of grammaticalization as languages in the speech modality do. But more importantly, this research shows that grammatical development may by-pass the general pathway from lexicon to grammar in that there is an additional pathway leading straight from manual or non-manual gesture to functional category (see "Evidence from signed languages"). In fact, a hypothesis to the effect that human language started out—primarily or exclusively—as a gestural system would be supported in particular by the following observations:

(a) Our closest relatives in the animal kingdom, the great apes, are good at gesturing but clearly less good at vocalization. Chimpanzees and other non-human primates, for example, have little voluntary control over their vocal signals but fairly good voluntary control over their manual gestures, and a pre-adaptation that was necessary for the emergence of modern spoken languages would simply have involved the extension of voluntary control from the hands to the vocal tract (Hurford 2003: 42–3). Note also that chimpanzee gestures are predominantly dyadic, that is, they involve reciprocal interactions between individuals,¹ whereas their vocalizations are generally not directed to specific others (Tomasello and Call 1997; Corballis 2002b: 167).

(b) Being more indexical and iconic than speech, gesturing provided a more readily available modality of comprehension and production than speech. In modern languages, certain concepts rely more heavily on gestural than on vocal expressions. Such concepts relate most of all to shape and spatial orientation; for example, when asked what a spiral is, people tend to resort to a manual demonstration (Corballis 2002b: 161), and a question for information on some location is likely to trigger a response that includes some pointing gesture.

(c) Gestures have been claimed to be implicitly syntactic (Armstrong, Stokoe, and Wilcox 1995), and if people are prevented from speaking and asked to communicate with gestures, their gestures tend to take on some syntactic format (Goldin-Meadow, McNeill, and Singleton 1996).

(d) That language evolved from manual gestures would also be supported by recent research on mirror neurons (Stamenov and Gallese 2002; Corballis 2002a, 2002b: 165).

To conclude, grammaticalization theory would be compatible with a gesturing hypothesis, or with a hypothesis arguing for a co-evolution of gesturing and vocalization, but so far lacks appropriate evidence to substantiate such a hypothesis—or else a hypothesis according to which language origin involved simultaneously gesturing and vocalization, followed by a gradual transition from the former to the latter (Auel 1980). Gesturing and signing provide promising lines of future research which could provide additional access to and new insights into the evolution of early language; but at the present stage of research, no conclusive answer to the present question seems possible.

(g) Can language genesis be related to behavior of non-human animals? This question has been controversially discussed in the literature on animal communication; nevertheless, we argued when relating our grammaticalization scenario (Table 7.1 in “Grammatical evolution Layers”) to the behavior to be found in non-human animals that there is a plausible link between the two: The language-related cognitive abilities observed in some animals can immediately be linked to early stages in the evolution of human language ("From non-language to language").

(h) Was language evolution abrupt or gradual? This question was the subject of "Did language arise abruptly?", where we showed that there is strong evidence to argue that language evolution proceeded gradually from one layer to another. Accordingly, a hypothesis in terms of an abrupt shift from non-language to language, or from early language to modern languages would seem implausible.

(i) Which is older—the lexicon or grammar? This question was the subject of "Lexicon before syntax", and the answer that we proposed is that there must have been some lexicon before morphology and syntax could evolve. Our findings on grammaticalization are thus in full agreement with the following observation made by Comrie (2000: 1000, 2002): If there is lexicon, then it seems that, at least in the presence of a community of potential speakers, language will develop, and will develop rapidly.

(j) What was the structure of language like when it first evolved? The answer to this question was given in "Grammatical evolution": Based on our scenario of evolution that was summarized in Table 7.1 in “Grammatical evolution Layers” we hypothesized that at the earliest stage, language consisted of one-word utterances being noun-like in character, followed at the second layer by the rise of noun–verb combinations allowing for mono-clausal verb-argument constructions.

(k) How did language change from its genesis to now? This question was answered in detail in "Grammatical evolution Layers”, where we hypothesized that language evolution must have proceeded incrementally from one-word utterances to propositional structures having some basic argument structure, subsequently to phrase structure, and eventually to clause subordination. And it also must have led from an isolating morphosyntax to one that was characterized by a gradual increase in agglutinating and inflectional properties.

(l) How long did it take to develop a structure that corresponds to what we find in modern languages? What we said about the preceding question also applies to this question. Grammaticalization theory offers a means to reconstruct relative chronologies, providing information, for example, on whether process X preceded or followed Y in time; but it does not allow for absolute dating of processes. Still, there is at least a partial answer that we volunteered in “Did language arise abruptly?”.

Since the development from concrete lexical forms to abstract functional categories, from basic sentences to complex structures of clause subordination, and from free forms to derivational and inflectional affixes must have taken considerable time to be accomplished, requiring a number of intermediate stages, the development from the beginning of human language to the structures characterizing modern languages is likely to have taken many millennia, perhaps tens of millennia.

(m) How did phonology evolve? The method that we applied was restricted to morphosyntax. This means that we had nothing to say about other grammatical phenomena, especially about phonology. Note that according to some, phonology was already there in early language, perhaps before syntax and other properties of language (Lieberman 1984, 1991, 1998; Carstairs-McCarthy 1999). Nevertheless, in the absence of any sound empirical data we follow others (e.g., Jackendoff 2002;² Bickerton 2005) in tentatively assuming that at the initial stages of early language there was no distinct phonological level and that it took some time for phonology to acquire a degree of internal structure—this is at least a position that is most readily compatible with findings on grammatical evolution. How exactly this may have happened is open to question.

But we suspect that the same general principles of evolution that we found in morphosyntax may also have been at work in phonology. For example, observations on phonological development suggest that some properties arose earlier in human languages than others (see Comrie 1992: 206–8, 2002; Newmeyer 2003: 72 for more details): (i) Since loss of earlier vowels or nasalizing consonants appears to be the ultimate origin of nasalized vowels, there must have been a stage in the evolution of human language in which a distinction between nasalized and non-nasalized vowels was nonexistent; thus, there may have been oral vowels and nasal consonants, but no nasalized vowels. (ii) Rounded front vowels can frequently be shown to go back to other vowels, which could mean that at some earlier stage there may have been no rounded front vowels. (iii) Since many tonal oppositions in language have been shown to have non-tonal origins, tone might not have been a distinguishing feature in early language. (iv) Since morphophonemic alternation is assumed to have developed out of a situation where there was no such alternation, it is likely that there also was an earlier situation without alternation, at least not of the many kinds that are found today. Such observations could suggest that at some earlier stage in the evolution of human languages there were no nasalized vowels, no rounded front vowels, no tone system, and no morphophonemic alternations. But since these are observations that are beyond the scope of grammaticalization theory, we will not attempt to draw any conclusions from them.

(n) How did the properties believed to be restricted to modern human languages arise, in particular syntax and the recursive use of language structures? We saw how syntax gradually evolved from one layer of grammaticalization to the next. But the main issue associated with this question is the rise of the recursive use of language structures. As we saw in “Early language Layers”, recursive structure was not present at the earliest layers of language evolution but must have emerged subsequently after the appearance of hierarchical noun phrase structure. It is only considerably later that recursive clause subordination came into being. Thus, our scenario suggests a development as sketched in (5).

The preceding discussion may have shown that there are answers to some of our questions, and at least partial answers to others. Still, there remain a number of questions that we had to leave unanswered. These questions have been the subject of a wealth of hypotheses and debates (see, e.g., the contributions in Hurford, Studdert-Kennedy, and Knight 1998; Givo̒n and Malle 2002; Wray 2002; Botha 2003a; Christiansen and Kirby 2003). But unfortunately, the methodology does not allow for any non-conjectural answers; these are issues that cannot be resolved without further research across disciplines.

¹ Reports on inter-animal gesturing indicate, however, that gesturing is not the most common means of communication, and when it occurs it tends to be limited to begging, taking, and embracing motions (Rumbaugh et al. 1978: 140).

² In Jackendoff’s scenario of evolution (2002: 244), the innovation of phonological structure is located at stage III, that is, after there was already a large lexicon and after the beginnings of syntax. While we are unable to assess this hypothesis, it is at least not incompatible with findings on grammaticalization.

اخر الاخبار

اخبار العتبة العباسية المقدسة

الهيأة العليا لإحياء التراث تناقش مع كلية الإمام الكاظم (عليه السلام) التحضيرات لمؤتمر ألفية الحوزة العلمية

قسم الصيانة يشارك في حملة التبرّع للشعبين اللبناني والإيراني

شرح دعاء أبي حمزة الثمالي.. إصدار جديد للعتبة العباسية المقدسة

العتبة العباسية المقدسة تعلن فتح منفذٍ جديد لاستلام التبرّعات داخل المرقد الطاهر دعماً للشعبين اللبناني والإيراني

المزيد

الآخبار الصحية

علاج طبيعي يخفف أعراض سكري الحمل

كم من الوقت يمكنك تخزين بقايا الطعام في الثلاجة بأمان؟

الوجه المظلم لـ"فيتامين B2" في معركة الجسم مع السرطان

فيروس بيئي يهزم "السالمونيلا" ويعزز سلامة الغذاء

المزيد

الآخبار التكنلوجية

خسائر غير مسبوقة للجليد في أيسلندا وسواحل المحيط الهادئ خلال 2025

رواد "ناسا" يستعدون لأول رحلة مأهولة نحو القمر منذ عقود

"أبل" تخطط لتحديثات ذكاء اصطناعي مهمة في مساعدها "سيري"

العلماء الروس يكتشفون طريقة تحسن فرص نجاح زراعة الأسنان

المزيد

مواضيع ذات صلة

The passive construction

The verb string: tense, aspect and mood

Grammatical functions and case

Early language Conclusions

Grammaticalization—a human faculty?

Did language arise abruptly?

The status of tree diagrams

Constituency: heads and dependents

Word classes

Core issues in grammar: comparing cognitive and generative accounts

Functional-typological approaches to grammar

Cognitive versus generative models