Previous work

The kinds of linguistic systems that we are concerned with here have been referred to with a range of different terms. For our purposes, a twofold distinction is of interest: On the one hand there are the modern languages, characterized by the following features: (a) they consist of the languages spoken today; (b) they are immediately accessible to reconstruction by means of established methods of historical linguistics; and (c) they relate to linguistic developments of roughly the last eight millennia. On the other hand there is what we will call early language,¹ which lacks all these features, that is, (a) it is not available today; (b) it is not accessible via orthodox historical methodology; (c) it is clearly older than 8,000 years and covers the timespan from the genesis of human language to the beginning of modern languages; and, consequently, (d) all we know about it remains of necessity hypothetical. Among the many works that have been written and the stimulating views that have been expressed it would be hard to single out any that deserve particular mention, or to provide an overview of this Weld of research that would do justice to all of them. We will therefore be satisfied in what follows with highlighting a few major orientations of scholarly thinking and themes that surface from this research; for a detailed and insightful discussion, see Johansson (2005).

A perhaps noteworthy observation is that quite a number of these works use what one may call integrating approaches. These approaches are based on the assumption that the more data on different and unrelated phenomena can be combined, the stronger is the hypothesis on language genesis and evolution that can be built; Givo̒n (2002a: 151) describes the logic of these approaches thus: ‘‘Many of the arguments for this are either conjectural, convergent or recapitulationist, so that they only make sense if taken in toto.’’ The kind of phenomena analyzed is largely the same from one author to another, including most of the following: (a) child language; (b) late second language acquisition; (c) pidgin languages; (d) agrammatic aphasics; (e) ‘‘Genie’’, the woman isolated from human contact from age three to thirteen (see Other restricted systems); (f) the behavior of non-human primates; and (g) regularities in linguistic change. Paradigm examples of such approaches are Bickerton (1990), Jackendoff (1999, 2002: 237), Comrie (2000), Calvin and Bickerton (2000), Givo̒n (2002a, 2002b, 2005), and Li and Hombert (2002). Givo̒n’s and Comrie’s approaches stand out against most others in that they take (g), that is, generalizations on linguistic diachrony, as an important parameter of reconstruction; this parameter is ignored by other authors, whose integrating approach can be characterized as being essentially achronic in nature.

The procedure adopted by Givo̒n and Comrie is mainly to look for common denominators among the structural properties of the phenomena listed above in order to propose a credible hypothesis on language genesis and/or language evolution. For example, Givo̒n (2002a: 151–2) concludes that grammatical concepts are more abstract than the lexicon and, given the overwhelming concreteness of primate and early childhood lexicons, it is unlikely that the evolution of the grammatical code could have preceded the evolution of at least some rudimentary well-coded lexicon in time.

Among the phenomena that we listed above, one has found particular attention, namely child language or ontogeny, for two different reasons.

First, the assumption implicit in some of the integrating and other approaches that have been proposed to reconstruct language evolution is that ontogeny recapitulates phylogeny—in other words, the way young children acquire language offers a possible analog of the way humans acquired early language. And second, it is argued by some authors that young children play a crucial role in language change, hence they also must have been the agents in language genesis and/or language evolution (e.g. Piattelli-Palmarini 1989). Neither of these assumptions is, however, really uncontroversial. While the ontogeny-recapitulates-phylogeny hypothesis appears to have considerable potential for reconstruction, it is not unproblematic; as has been demonstrated by Slobin (2002), there is substantial counter-evidence to this hypothesis. And that young children were instrumentally involved in language genesis is also an assumption that is in need of reconsideration (see Who were the creators of early language?).

Models taken from evolutionary biology While there is a long and respect able tradition of research in evolutionary biology, there is no corresponding Weld of evolutionary linguistics. It is therefore not surprising that students of language evolution have drawn on models from biology in framing or justifying their hypotheses or for proposing explanatory accounts. A number of controversies on what are essentially linguistic issues have been based on biological notions and, accordingly, the hypotheses proposed have been shaped by the respective biological models (see Botha 2003a for detailed discussion).

One major line of research is based on Charles Darwin’s ([1859] 1964) concept of adaptation, according to which a fitness-enhancing character is shaped or built for its present function by natural selection. In the same way as a complex design like the eye of a vertebrate can be accounted for by adaptation and gradual evolution, it has been argued since Pinker and Bloom (1990) that the complex design of human language can be explained as a biological phenomenon in the Darwinian sense, having arisen via a series of adaptive processes (Pinker 2003; Jackendoff and Pinker 2005; see also Haspelmath 1999c; Croft 2000; Mufwene 2001; Bierwisch 2001; Givo̒n 2002a, 2005). that have been invoked. Thus, it is argued that language would have allowed our ancestors to acquire information and pass it on far faster than biological evolution could achieve, giving them a decisive advantage in competition with other species. But whereas many agree that ‘‘language cannot have come cheap’’² there is no real consensus on what could have been the reason for all the expensive changes that evolution has brought about in order to give us speech. On one view, let us call it the ‘‘hunting’’ view, early man was a great hunter, who needed to communicate plans for herding prey or trapping them in particular places. Our ancestors needed to communicate about places to hunt, new sorts of traps, locations of water, good caves, techniques for making tools, and ways to make and keep fire. The bottom line is: Men needed to speak in order to hunt better. But there is also a ‘‘foraging’’ view on which, at the dawn of human language, communication was necessary in order to exchange information about locations, the nutritional value of available foodstuffs, and the safety of available foodstuffs. On the ‘‘narrative’’ view, language evolved to enable the exchange of stories about the supernatural or the tribe’s origins. According to the ‘‘Machiavellian Intelligence’’ view, social primates are believed to understand matters such as alliances, familial relationships, dominance hierarchies, and the trustworthiness of individual members of the group. The assumption therefore is that our hominid ancestors were social animals, that they could recognize and compare different social relationships, and that they could respond appropriately.

Social life is complex: it requires a lot of brain power to remember who is who, who has done what to whom, etc., and language therefore is said to have emerged in order to enable and maintain the complexity of social life, and on the ‘‘gossip’’ view, language evolved to allow us to gossip (Dunbar 1996). Dunbar argues that most of our talking is gossip, which enables us to cement and maintain our human relationships, in short: Gossip does for humans what grooming does form any other primates, language makes it possible for us to gossip. It acts as ‘‘a cheap and ultra-efficient form of grooming’’ (Dunbar 1996: 97; see also Dunbar 2006): We can talk to more than one person at once, pass on information about cheaters and scoundrels, and tell stories about who makes a reliable friend. Central to the ‘‘memetic’’ view is the notion of meme as an element of culture that may be considered to be passed on by non-genetic means, especially imitation (Blackmore 1999). Blackmore considers the meme to be an entity that plays the role of gene in the transmission of words, ideas, faiths, manner isms, fashions. Humans have the unique ability to imitate, and so to copy from one another ideas, habits, skills, behaviors, inventions, songs, stories. Memes, like genes, are assumed to be selfish replicators; they compete to find space in our minds and cultures, for the sake of their own replication. The standpoint taken in memetic theory is that language evolved for the spread of memes: ‘‘When the environment changes, a species that can speak, and pass on new ways of copying, can adapt faster than one that can adapt only by genetic change’’ (Blackmore 1999: 99; see above).

The main problem associated with the above approaches having a Darwinian adaptation orientation concerns appropriate empirical evidence on the nature of selection pressures that can be held responsible for adaptation. Whereas evolutionary biologists have come up with sufficiently detailed reconstructions of the evolution of organisms, there is so far no really convincing hypothesis on the nature of the adaptive processes that ultimately led to the emergence of modern languages in all their phonological, morphological, syntactic, and semantic complexity (see Johansson 2002: 94–108, 2005).

An alternative line of research uses exaptation as an explanatory construct. Gould and Vrba (1982: 6–7) proposed the notion exaptation for biological characteristics that enhance fitness in their present role (or effect) but were not built by natural selection for this role. On this view, certain fundamental features of language originated like birds’ wings and tail feathers, which evolved initially as an adaptation for thermoregulation and were later exapted for an entirely different function, namely for flight (e.g. Lieberman 1991; Gould 1991; Wilkins and Wakefield 1995, 1996; Carstairs-McCarthy 1999; Calvin and Bickerton 2000).

A related major line of research relies on the notion of spandrel cooptation, a variety of exaptation where characteristics that did not originate by the direct action of natural selection were later co-opted for their current utility or function. Characteristics belonging to this subcategory have been called ‘‘spandrels’’, a term borrowed from architecture: Spandrels are forms and spaces that arise as necessary by-products of another decision in design rather than as adaptations for direct utility in themselves (Gould 1997: 10751). On this view, the structures governing language are biological traits which originally served no function at all but acquired their present function via exaptation or ‘‘reappropriation’’ (cf. Piattelli-Palmarini 1989: 19; Wilkins and Wakefield 1995).

A problem with the last two kinds of hypotheses on language evolution is that they claim that something has or may have happened, but they do not show how it happened—that is, they do not offer any coherent account of the process leading from non-language to early language, or from early language to modern languages. While biologists have come up with fairy appropriate descriptions of exaptive processes, it remains unclear how exactly exaptation may have proceeded in the evolution of language structure.

As these remarks suggest, there is an enormous diversity in views that have been held in understanding language evolution, and it comes as no surprise that some of these views have given rise to controversies. In the following paragraphs we highlight some of these controversies that are particularly relevant for an understanding of the subject matter covered here.

From ape communication to language One controversy relates to the question of whether language can immediately be derived from forms of communication or cognitive abilities to be found in non-human animals, most of all in apes, or whether it is a phenomenon sui generis that has no homolog in animal behavior, having evolved in a discontinuous fashion without any links to pre-human forms of communication and/or cognition. In accordance with the former position, Jackendoff (2002: 236) argues that one can reconstruct from modern human language a sequence of distinct innovations over primate calls, with each being an improvement in communicative expressiveness and precision. The latter position, emphasizing the uniqueness of human languages, asserts that specific properties of modern languages, most of all the syntactic mechanism of recursion, do not appear to have any homolog in non-human animals (Hauser, Chomsky, and Fitch 2002). But there are also compromising positions that try to reconcile these contrasting views. A more differential perspective of continuity is propagated in particular by Aitchison (1998: 19–22): Distinguishing three constituents of language, namely (a) auditory mechanisms, (b) articulatory mechanisms, and (c) the brain, she maintains that continuity was strong in the case of (a), weak in the case of (c), while (b) shows the greatest discontinuity.

Discontinuity vs. continuity Irrespective of whether a bridge can be built between non-human animal communication and human language, there remains another question, namely whether the rise and early evolution of human language was gradual or abrupt (see Botha 2003a: 36–41 for discussion); let us refer to the two main claims that have been made on this issue as the gradualist hypothesis and the leap (or discontinuist) hypothesis, respectively. Proponents of both hypotheses draw on evidence from evolutionary biology and on biological notions such as adaptation, mutation, and exaptation in search of support for their respective positions.

The gradualist hypothesis³ is strongly associated with, though not restricted to, the Darwinian (or Neo-Darwinian) paradigm of natural selection and adaptation. Arguments in support of the gradualist hypothesis tend to emphasize, first, that language evolution should not be viewed as being drastically different from other linguistic, socio-cultural, and biological phenomena. And second, on account of their enormous complexity, human languages—just like the vertebrate eye—cannot have arisen overnight but must have involved a series of developments, with one incrementally building on the other (e.g. Pinker and Bloom 1990; Newmeyer 1998b; Bichakjian 1999; Jackendoff 1999, 2002; Givo̒n 2002a, 2002b, 2005; Heine and Kuteva 2002a; Tomasello 2003a). For example, Newmeyer (1998b: 311) holds that syntax is made up of various subsystems each of which is governed by a distinct set of principles which cannot be derived from a more fundamental principle or property. He concludes that these principles must have evolved in an incremental way, thus suggesting a gradualist scenario.

The leap hypothesis capitalizes on what is argued to be the unique nature of language (Kirby 2002: 173). Supporters of this hypothesis include Chomsky (1988) and Gould (1997), who argue that language may have evolved all at once as the product of a kind of macromutation (see Pinker 2003: 25), and roughly the same position is maintained by Bickerton (1990, 1998), for whom there was a single mutation, coincident with the transition from Homo erectus to Homo sapiens, that created true language out of what he calls ‘‘protolanguage.’’ The major linguistic consequence of this mutation was the imposition of recursive hierarchical structure on pre-existing thematic structure, transforming ‘‘protolanguage’’ in one swoop to true modern human language:

the linkage of theta analysis with other elements involved in protolanguage would not merely have put in place the basic structure of syntax, but would also have led directly to a cascade of consequences that would, in one rapid and continuous sequence, have transformed protolanguage into language substantially as we know it today. (Bickerton 1998: 41)

Evidence for this hypothesis is seen in particular in the syntax of human languages whose nature is claimed to preclude the possibility of a gradual evolution (Berwick 1998). The following arguments in particular have been volunteered in support of the leap hypothesis or—more appropriately—against the gradualist hypothesis (see Botha 2003a: 79V.): First, there is no evidence of a gradual accumulation of linguistic capabilities over a long period. If indeed there had been a gradual process, then either we should find some level of ‘‘stable syntax’’ somewhere between early language and modern languages—showing up in atypical manifestations of human languages (such as aphasic and dysphasic syndromes), cryptolects, stages of first and second language acquisition—or there should be synchronic varieties of language bearing witness of intermediate stages (Bickerton 1998: 354–7; Crow 2002). Second, the syntax of modern languages is such that it could not have evolved in an incremental fashion: Berwick (1998: 338–9), for example, argues that many syntactic relations and constraints, such as basic skeletal tree structures, movement rules, etc., can be derived from ‘‘Merge’’, a combinatorial operation used in generative grammar for the derivation of sentences. Accordingly, it is argued, if there had been evolution of syntax, then it must have involved a kind of leap from non-combinatorial syntax to ‘‘Merge.’’

Both hypotheses highlight relevant issues of language evolution. But there is a difference. Whereas work based on the gradualist hypothesis has come up with scenarios on how the transition from non-language to modern languages can be accounted for (e.g. Jackendoff 2002; Heine and Kuteva 2002b; Davidson 2003), there appear to be no appropriate scenarios in support of the leap hypothesis. It remains largely unclear how, all of a sudden, the complex morphological and syntactic structures characterizing modern languages came into being. We will return to this discussion in “Did language arise abruptly?”.

A language-specific faculty? There is a wide range of further issues that have occupied students of language genesis and evolution over the past decades. The reader is referred in particular to the many contributions to be found in the volumes edited by Gibson and Ingold (1993), Hurford, Studdert-Kennedy and Knight (1998), Givo̒n and Malle (2002), Wray (2002), or Christiansen and Kirby (2003), Tallerman (2005), and the many books that are available (e.g. Aitchison 1996; Bickerton 1990; Calvin and Bickerton 2000; Botha 2003a; Givo ´n 2002a, 2005; Johansson 2005). Mention should be made in particular of one line of recent work which relates language and language evolution to animal communication systems on the one hand, and to non-language-specific human cognitive abilities on the other. Modern languages have a number of properties that contrast sharply with anything that we find in non-human primates and other animals, and that have puzzled researchers trying to establish links between forms of communication to be observed in apes and what we find in languages today. Properties that have been mentioned in particular are multi-propositional discourse coherence, spatially or temporally displaced reference, syntactic displacement, that is, the ability to move constituents from their natural argument positions and place them in other slots in the sentence, locality constraints which prevent displacement from acting over unbounded domains, or structural dependence (Hauser and Fitch 2003; Givo ´n 2005; see also the ‘‘design features’’ of Hockett 1960).

But there is one property that has found particular attention. Hauser, Chomsky and Fitch (2002) propose distinguishing between a faculty of language in the broad sense (FLB) and a faculty of language in the narrow sense (FLN). According to these authors, most, if not all, of FLB is based on mechanisms shared with non-human animals, while FLN is specific to humans and human languages; FLN, these authors argue, comprises only the computational mechanisms of recursion as they appear in narrow syntax and the mappings to the interfaces. In short, recursion, that is, the embedding of a constituent within another constituent of the same type,⁴ is claimed to be essentially the only uniquely human component of the faculty of language, being specific both to language and to humans. So how did this property arise in human languages?

New directions Some of the problems that we dealt with above may perhaps be solved by future research. In the course of the last decades a number of new lines of research have been developed that might shed light on problems that previously were not accessible via empirical research. The following remarks are far from presenting an exhaustive account of all the new directions that have surfaced in the recent literature; rather, we are restricted to mentioning a few salient research areas (for more detailed discussion, see Johansson 2005).

One promising direction of research is provided by work on computer simulation. Common to many of the approaches that have been devised is the theory that there are agents, intending to communicate and produce a self-generated and self-organized language-like system (e.g. Batali 1998; Steels et al. 2002; Steels 2003; Steels and Belpaeme 2005; Briscoe 2002; Tonkes and Wiles 2002; Kirby 2000, 2002).

Another promising line of research can be seen in the study of pre historic human behavior as it manifests itself in particular in ritual burials, sea-crossings, and artefacts such as rock engravings, cave paintings, statuettes, and other pieces of art, where creativity accelerated dramatically roughly between 85,000 and 50,000 years ago (e.g. Davidson and Noble 1993; McBrearty and Brooks 2000; Klein and Edgar 2002; Mellars 2006; d’Errico et al. 2005; Henshilwood 2006; Conard 2006; Wurz 2006). Abstract geometrical designs, perforated shell beads, bone tools, and other products of early human behavior that were found in southern African sites—for example Blombos Cave, Klasies River, and Diepkloof—are taken by some to provide strong evidence for the presence of language during that period. So far, however, no compelling correlations between these findings and language evolution have been established (cf. Coleman 2006).

Furthermore, there are Welds such as neurobiology and genetics, including epigenetics, which have a large potential for providing new insights into how early language arose and evolved. The discovery of mirror neurons in the brains of macaque monkeys allow us to view basic abilities required for communication, such as imitation and speech control, in a new light, and research on FOXP2, a gene on Chromosome 7, might lead to a better understanding, for example, of the contribution made by nonhuman primates and of possible selection pressures that could have influenced language genesis. Both neurology and genetics are dynamic research areas and it is hard to predict how they will enrich our knowledge on early language in the years to come.

Finally, another line of work can be seen in the search for ‘‘linguistic fossils’’. This search, which has its parallels, for example, in observations on archaic features in the physical structure of present-day organisms, is concerned on the one hand with the study of restricted linguistic systems (Botha 2003b, 2005, 2005/6); on the other hand, it is based on the claim that certain design features of modern languages exhibit an evolutionarily ‘‘more primitive character’’, or resemble fossils of earlier evolutionary stages (Fo̒nagy 1988; Bickerton 1990; JackendoV 1999: 276, 2002: 264). Using ‘‘fossil’’ phenomena for reconstruction is not new in linguistics, it is essentially part of the method of internal reconstruction and other diachronic approaches (Givo ´n 2000), and it also has some potential for the reconstruction of early language. A problem that needs further attention— in this case as in other cases that we discussed here —concerns the basis of comparison between two kinds of phenomena. For example, when Jackendoff (2002: 264) hypothesizes that a certain design feature X of modern languages resembles (or is related to) some feature Y of early language then this raises the question of what one actually knows about Y or—more precisely—what the methodological basis for one’s hypothesis on Y is. Jackendoff does not elaborate on this issue, other than alluding to defective lexical items, items that have no syntax, etc., or noting that some of the defective lexical items ‘‘have almost the flavor of primate alarm calls,’’ etc. In other words: Can we relate X and Y meaningfully to one another without having an appropriate theory that allows us to set up Y?

These and many other new directions of research suggest that, after a history of more than three centuries, work on language genesis and evolution, while still faced with fundamental problems, might now be entering a new phase—one that is determined more by empirical observations than speculation, and whose outcome is promising, though hard to predict.

¹ Early language must not be confused with ‘‘protolanguage’’ (Bickerton 1990), a term that will not be used in this work except when referring to works where it appears.

² The whole of our vocal apparatus had to be evolved, which meant (a) complex changes in the neck, mouth, and throat, and (b) the impossibility to drink and breathe at the same time, increasing the risk of choking.

³ Bickerton (2005) refers to this hypothesis as ‘‘genre continuism.’’

⁴ This is only one way in which the term recursion is or has been used.

اخر الاخبار

اخبار العتبة العباسية المقدسة

الهيأة العليا لإحياء التراث تقيم حفلًا لتكريم الفائزين في مسابقة القصة القصيرة عن فاجعة الطف الأليمة

العتبة العباسية المقدسة تستقبل المشاركين في النسخة الثالثة من جائزة العميد الدولية القرآنية

العتبة العباسية المقدسة تطلق فعاليات ملتقى النورين القرآني الرمضاني الخامس في بابل

قسم الشؤون الفكرية ينظّم دورة تخصصية في فهرسة وصيانة المخطوطات بجامعة ديالى

المزيد

الآخبار الصحية

للطاقة وحماية القلب.. ما "التوقيت الذهبي" لأكل المكسرات؟

دراسة تكشف فائدة القيلولة لمدة 45 دقيقة

اختيار الزبادي الأنسب.. قرار يحدد مسار صحتك

هل يعدّ الصيام المتقطع حلا سحريا لخسارة الوزن؟.. دراسة تجيب

المزيد

الآخبار التكنلوجية

"علي بابا" تكشف عن نموذج جديد للذكاء الاصطناعي

اختراق علمي جديد.. استشعار ضوئي يرصد "دلالات السرطان"

وكالة الإمارات للفضاء تعلن تمديد مهمة "مسبار الأمل" حتى 2028

ظاهرة مذهلة .. شهدها 2 بالمئة من سكان العالم فقط

المزيد

مواضيع ذات صلة

Determiners and quantifiers

Nominal grounding predications

Class schemas

Atemporal relations

Temporal relations: verbs

Assumptions

Temporal versus atemporal relations

Nominal versus relational predications

Abstractions

Expansibility and contractibility versus replicability

Homogeneity versus heterogeneity

Bounding