The Role Of Computer Simulation in the Evolution of Language Debate BA (Hons) essay for Ling 40? - Evolution of Language. MrAnton Melser 2002Anton Melser G System Philosophy Introduction Computers and computer technologies are now ubiquitous in academia and, indeed, in most areas of modern Western life. While some hold reservations about the true contribution made by computers, many, if not most, believe the limitations for us now are not with computers but with our imagination. Universities and technical institutes have embraced computer technologies with enthusiasm and many faculties now will not accept essays or assignments if they have not been put through a word-processor. Registration, course information and, increasingly, course content are available through the Internet. The internet has a wealth of information and resources (a good deal of the material for this essay was downloaded from the internet) and in reality it is only cost which inhibits universal access. The real revolution for academia, however, lies not in the mere exchange of words and pictures but in the construction and exploration of new and exciting artificial worlds. It is the relevance and contribution of these artificial realities or simulations to the Evolution of Language (henceforth EOL) debate that is the topic of this essay. In the first part of this essay I take a look at what a computer simulationIn this essay I use the terms “simulation”, “computer simulation”, “computational model” and “computer model” interchangeably. This is not unproblematic, but simplifies things greatly and is not overly detrimental to the points made. is and in particular what a simulation of the origins and evolution of a language/communication system is. I ask why we would want to build such simulations and look at what modelling opens up to us. Some recent work is reviewed. The essay then moves on to look at tendencies in the field and shifts that appear to be taking place. In the final section I take a look at some of the theoretical and practical problems that surround simulation work in EOL studies and look at how the work fits in to the overall research enterprise. Simulation and Complex Systems What is a computational model? Roughly speaking a computational model can be described as a numeric approximation of a situation or system instantiated in some form on a computational device. In practise this approximation will usually be a simplification incorporating the salient features of some real world physical or social situation upon which various functions or mappings are performed. These functions represent other phenomena or processes which are known to take place in the real world. These mappings are run on computers and, hopefully, we end up knowing something about how the real world phenomena interact - something more than we would have been able to find out with pen and paper. It is important to note that everything is represented (as all things on modern computers are) necessarily as a finite bit string, a point I will return to briefly (see Flake, 2001 and references therein for more on computation). What can computer models offer EOL research? The attraction of computer modelling is obvious. More and more cheaply, models can be built to simulate just about anything, from the physics of elementary particles to the art of music composition, from nuclear chemistry to the functioning of national economies, from evolution to the world of Shakespeare. Simulations are lightning quick in comparison to many physical, biological and social experimentation methods. A complicated simulation might take as little as a few minutes on today's machines. Simulations never suffer from dirty test tubes or interviewer bias and can be performed at any time of the day or night, rain or shine. Another hugely attractive aspect for ecologists, nuclear researchers and social scientists alike is the freedom to run experiments unthinkable in the real world. A meteorologist can't test the effects on the upper atmosphere of letting off 100-megaton nuclear bombs, and a cognitive scientist can't deprive a child of all human contact to test hypotheses of intellectual development - all of this and more, however, is but an "Enter" away in the artificial realities of the computer. The key question I look at in this essay is - given there is a world of promise, what has been delivered to EOL researchers? The EOL discussion is a perfect candidate for simulation experiments and for well over a decade now research in the development of communication systems, or various aspects of them, has flourished (see below for references). Because language developed in the distant past (all agree more than 60,000 years ago) there is no way to directly gather evidence, at least for its development in our own species. Now also (at least in the West) all experiments with human beings need to go through an often lengthy approval process with ethics committees, and issues of privacy and representative sampling (i.e., the first-year psychology student problem) often detract from results. However, language (it is thought by most) is unique to Homo Sapiens and so experimentation with other species is often dismissed as totally irrelevant (see Wallman, 1992 or Taylor, 1997). On top of this, research with other species can be horrendously expensive and take years to get results from (as with the work of Rumbaugh and Savage-Rumbaugh, see Savage-Rumbaugh and Lewin, 1994). This often makes such research impracticable. What phenomena have been modelled in the EOL field? Simulation work in the field can be roughly divided into work on syntax (e.g., Batali, 1994, 1998; Briscoe, 1998; Gmytrasiewicz and Gopal, 2000; Hurford, 2000; Kirby, 1997, 2000; Kirby and Hurford, 2001; Steels, 1997; Tonkes and Wiles, 2002; to name but a fraction), the lexicon (e.g., Kaplan, 1998; Oliphant, 1996; Smith, 2001; Steels, 1998, Van Looveren, 2001; Vogt, 2001; as above) and the less common work on phonology/phonetics (e.g., de Boer, 2000, 2001; Glotin and Laboissiere, 1996; Berrah, 1998 (cited in de Boer, 2000)). Though it would be impossible to give an accurate account of how each simulation was run, it is possible to review some of the main mechanisms. The basic format of many of the simulations (see Steels, 2000) can be described in the following way: A group (sometimes as small as two) of agents is put together and engages in controlled interactions or language games. The agents usually have no ability to communicate initially, at least in the phenomenon being modelled and, after a certain number of language games, an ability and/or language develops with which enables the agents communicate. The extent of knowledge and abilities assumed varies (see below for discussion). The structure of the agents varies but they are commonly represented as neural networks (e.g., Batali, 1998; Tonkes and Wiles, 2002) or simply as a collection of numeric weights (de Boer, 2001) or physically instantiated as robots (Steels, Kaplan et al, 2002) with collections of numeric weights. Steels (2000:1) describes the methodology thus:
The basic idea is that a community of language users (further called agents) can be viewed as a complex adaptive system which collectively solves the problem of developing a shared communication system. To do so, the community must reach an agreement on a repertoire of forms (a sound system in the case of spoken language), a repertoire of meanings (the conceptualisations of reality), and a repertoire of form-meaning pairs (the lexicon and grammar).
In the above quote Steels mentions the notion of a complex adaptive system. A full analysis of the notion of complex adaptive system is out of the question here and the reader is directed to Flake (2001) or Stonier and Xing (1995) for a more substantial treatment. The following excerpts give us an indication of what a complex adaptive system entails:
Complex Systems are things that consist of many similar and simple parts. Often the underlying behavior of any of the parts is easily understood, while the behavior of the system as a whole defies explanation. ... By changing the type and form or interactions that exist among the parts of a complex system, the type of global behavior can be varied such that the complex system as a whole can be globally goal-seeking while only local information is passed around by the parts. This means that a collective form of computation can take place without an explicit global algorithm. (Flake, 2001:229)
The analysis of systems with nonlinear interactions among system components dominates many aspects of current research. Such systems with interesting emergent behaviour are referred to as complex systems. Those complex systems with the additional property that their primitive components can change specification, or evolve, over time, are often called complex adaptive systems (CAS). The basis of adaptation rests on the premise that there is some condition of operation or performance which is better than any other. Moreover, to be called adaptive, self-organising features must exist in the system to enable performance to be optimised. (Stonier and Xing, 1995:Introductionhttp://www.csu.edu.au/ci/vol02/intro.html)
The basic idea is that groups of "agents" or units are designed to exhibit only very limited behaviours. Out of prolonged interaction a language or language ability develops. Tendencies - from genes to ants A lot of work in the late 80's and early 90's looked at the genetic development of groups' of individuals (agentsLike most authors working on simulations, I will be intentionally vague ( at least in this part of the essay) in my terminology with “individuals”, “societies” or their analogues “artificial agents and “artificial societies”.) communicative abilities using geneticsee Mitchell (1998), and in particular section 3, for an general introduction to research in this area. algorithms. It is no coincidence that it was around this time that Pinker and Bloom (1990) burst into flower. It is probably also true that at this time it was widely accepted among linguists/cognitive scientists that there existed some sort of language-specific processing machinery in the brain This, of course, goes back to the 19th century with Broca and Wernicke but it is probably fair to say that, until the development of modern neuroscience and psycholinguistics, this was largely speculative., as well as a LAD in more than the abstract sense. This approach was also supported by the belief in humans' unique status as owners of a combinatorially rich and unbounded system of communication. There seemed to be only one mechanism for the development of such a system - evolution by natural selection, and consequently it was to genes and genetic mechanisms where we were to look (and so to model). The models created simulate the development of language specific capabilities in the species. Examples of this type of simulation are Hurford (1989, 1991), Werner and Dyer (1991), Lucas (1994) and Hashimoto and Ikegami (1995). Since that time, however, and possibly thanks in large part to the work and prestige of the scholar Luc Steels, in the past decade there has been a move away from genetic approaches - away from sexual selection of/for abilities/capacities towards a self-organisation approach. The mid-90's saw a good deal of what could be termed genetic assimilation (Baldwinian evolution) modelling approaches. These incorporated genetic aspects and self-organisation aspects. See Batali (1994), Oliphant (1999), Livingstone and Fyfe (2000), Kirby (1998), Kirby and Hurford (1997), Briscoe (1998). Recently, though, recourse to genetic explanation of any kind has been explicitly rejected in favour of wholly self-organisational approaches (see Steels, 2000), see de Boer (2000, 2001), Hutchins and Hazlehurst (1995), Hurford (2000), Kirby (2000), Batali (1998), Steels (1998), Vogt (2001) among many others. With the shift away from genetic selection to a more consensus-based mechanism there has been a shift from focus on the individual's (agent's) abilities to a focus on the development of the actual communication system (language) itself. An good example here is the "Glossogenetic" approach of Kirby and Hurford (2001, inter alia). This refocussing follows directly from the fact that the agents "capabilities" don't change over time, only the values of certain parameters. The actual values taken by these parameters are usually arbitrary and can change even after a stable system has been obtained. The agents have very little or no language-specific machinery and there is no change in agent architecture as such. There are a number of issues with this move, some of which are discussed below. These are certainly only tendencies, however, and contradicting examples could easily be found. For example, Cangelosi (2001) is still heavily genetic in approach, and so on. What does seem significant about this is the theoretical shift implied. Many in the field openly reject the Chomskyan innatist view of language ability (e.g., de Boer, 2000; Steels, 2000). If this trend continues and takes hold among linguists we may see a move back to the more learning-based theories of the pre-Chomskyan era. There is evidently a wish to explain as much as possible in non-genetic terms. Having said this, however, most of the scholars would not in any way attempt to deny the importance of genetic factors. I look at this in more detail below. Problems I now turn a critical eye on simulation work to ask the question. How has it helped us in solving the enigma of the origin of language? The first part of the section deals with problems with a totally self-organisational approach and the last part deals with simulation issues in general. One of the key aspects of self-organising systems is the bottom-up development of structure. One of the prototypical examples of this phenomenon is the formation of near optimal paths to a food source in a species of ant, Linapithema humile (see Bonabeau and Theraulaz, 2000 for a fuller description). Though it is clear that an ant has no concept of optimising the time and energy spent getting to a food source, in spite of this, the optimal path is (often) obtained. The mechanism works in the following way. On returning to the nest from a plentiful food source an ant drops pheromone. Other ants are attracted to this pheromone trail, and so to the food source, but with a very weak trail, the ants will initially take very many different paths back to the nest. The number of ants increases and so does the amount of pheromone between the nest and the food source. The paths that are most direct to travel will, on average, be those ones that are reinforced the most and will, after a time, begin to have significantly more pheromone. Eventually ants will begin to use these paths exclusively. So it appears that an optimal path, sometimes a very long one through even the immensely varied jungle terrain is reached. This sort of result looks almost intelligent, though once we look at the mechanism we see clearly it is not. It is in this way that an increasing number of researchers wish to investigate the origins and evolution of language. In many ways, however, this is very strong claim. If it is an advantage to be a successful communicator (and most take this point for granted, though there are powerful arguments against this position, cf. Power, 2000; Dessalles, 2000) and all that is required for this are a few basic all-purpose learning mechanisms and a modicum of habitual association then how is it that only humans have language? We know even lower primates (e.g., Vervet monkeys, Cheney and Seyfarth, 1990) are capable of maintaining a digital lexicon, and capable of storing and processing intricate tapestries of social information. And what of the chimpanzees of Savage-Rumbaugh, who have shown the capability of acquiring at least a rudimentary form of what we might call a "protolanguage" (in the sense of Bickerton, 1990)? We must ask why this has not developed into full-blown language if the principles of self-organisation obtain in the real world. The urge to over interpret results should be resisted. If it is a mistake to anthropomorphise the behaviour of the higher primates, who share so much with us in the real world, then it is surely a faux-pas of the worst kind to read anything into rudimentary simulations. It is here we see a key problem in taking the self-organisation approach. Artificial agents are not even ants, let alone vervets or proto-humans. While this approach appears to have been fruitfully applied in the economic sphere (see Arthur et al., 1997 for examples) the fact that many aspects, such as differential advantage in communicating between the sexes - an aspect some claim to be of vital importance (e.g., Power, 2000) - are utterly ignored. It seems to me that, in essence, what is happening is that the lessons learnt over the past decade or two in the EOL debate are abandoned - for the sake of constructing manageable models with clear-cut results. Though this might bring simulation into the mainstream, it seems an intellectually questionable move. It was noted above that there is a distinct tendency in current simulation experiments to deny any role within the model to genetic evolution. The rationale behind the move seems innocuous enough - the less we need to explain in terms of physical genetic evolution the better. We obviously have no way of knowing the exact phenotypic/genotypic progression in the hominid line, let alone the sequence of specific selective pressures that were present. Therefore, all structure that can be explained in terms of the interaction between low-level elements decreases the load on genetic explanations and should strengthen our theory. However, to ignore sexual (genetic) selection entirely is to fly in the face of established facts about how biological organisms interact with their environments over time. It also seems highly unlikely that the physical prerequisites (short-term and long-term memory; vocal apparatus - serial gesture coordination, lowered larynx, etc) could have developed in the precise ways they did without any selective pressure. While it may be conceivable that general purpose mechanisms could be responsible for the development of 5-, 10-, or even, 50-item lexicons, modern humans have lexicons of tens of thousands, sometimes hundreds of thousands of items. The hardware necessary for this sort of processing simply could not have been left lying idle until one day the group suddenly started speaking to each other. By the same token, if we prefer the theory that the necessary hardware was already instated (as in the "sudden-change" theories e.g., Bickerton, 1990) and was simply coopted for speaking then at the very least it needs to be shown that the increased processing load was bearable. This "monster-mutation" theory of Bickerton is losing favour in the EOL debate however, and even Bickerton seems to be tempering his stance (see Bickerton, 2002). At the very least, the assumption of physical prerequisites is not unproblematic. There are a number of language-specific adaptations, such as our sound-producing capabilities which have no analogue in the higher primate world. Why would humans develop sometimes dangerous adaptations, such as a lowered larynx, if not under pressure to increase the range of sounds in an already-present repertoire? None of these issues are adequately dealt with by the simulationists (though see the work of Daniel Livingstone and Colin Fyfe (e.g., 2000) for an attempt to address physiological issues). It is not that none of the scholars are trying to address these issues - many are - but too often, it seems, we are told "future models will include" this or that. Luc Steels, for example, has done a good deal of recent work on robots, which he feels are more realistic, as they will introduce physical and temporal constraints not present in completely "soft" simulations. I wholeheartedly agree that this is a positive more, but it forces us to ask whether even the most sophisticated "soft" simulation is lacking in the vital ingredient of physical context/situation. While the robot's world is not the savannah of East Africa, it is considerably more open to the sort of indeterminacies that would have been present for our distant ancestors. Surely, also, this indeterminacy played a role in the development of language. Just as surely, it is lacking from the sterile cyber-worlds. We might go even further, however, and ask whether this sort of modelling is appropriate for modern, rational, free agents such as humans. Simulationists assume that the essential properties of human rationality, action and agency have been formalised. This point leads into the issue of the computability of real world phenomena. There is an implicit assumption that the relevant aspects of a situation can all be translated into a finite bit string, and that this translation is unbiased and complete. While this is an area of debate in computer science and well out of the scope of this essay, suffice it to say that assuming that continuous, real-world phenomena can be represented by finite binary bit strings is not uncontentious. There is even a school of thought (social constructionism) that would claim that all of these things are by nature incomputable because their interpretation is "up for grabs" (see Shotter, 1993). Can we really model just one aspect of human communication? (e.g., vowel systems?). Doing this suggests that these areas are either separate/independent from other parts of the system or, alternatively, that we know all of the other relevant factors and can accurately determine how and how much they affect the functioning and development of the whole. As is evidenced in the variety of theories to be found in any of the main texts coming out of the series of Evolution of Language conferences (Hurford et al, 1998, 2000; Wray, 2002), there is certainly no consensus on what the important factors were, nor how they might have interrelated. MacLennan and Burghardt (1993) make the following comment, which is clearly germane:
In a simulation, an attempt is made to imitate in a computer or other modeling system the salient aspects of a system that exists, at least potentially, in the real world. The design of a simulation is heavily theory-laden and necessarily highly selective. This is true even for models based on current theoretical and empirical understanding of the phenomena being studied. For out of the multitude of features in the natural situation, only a small fraction can be selected for modeling. This is the Achilles' heel of simulation, for an inappropriate selection vitiates the relevance of the model. This problem is especially critical in ethology, because animals respond so sensitively to their environments that it is often unclear whether a feature is relevant or not. Indeed, whether a simulation and its underlying assumptions is considered useful or valid is often based on how robustly it matches our expectations.
Conclusions Perhaps simulation is not the panacea once hoped for. If we are to build realistic models - and surely we must to attempt to tackle social or cultural questions - then our models will not be simple, and will take considerable time and expense to construct. At the end of the day, a simulationist can always respond with the claim that, without the sort of groundwork and bank of experience working with low-level, oversimplified models, we will never be able to model more realistic worlds. In the final analysis I would have to agree. Preliminary low-level modelling is necessary. As a linguist and computer scientist, however, I would urge scepticism concerning any results claimed from this early simulation work. It is not clear that the models take enough complexity into account to discover anything we could not have readily predicted from initial conditions. Simulations will develop, though, and it seems only a matter of time before simulation becomes a truly necessary investigative tool. I do believe we will need patience though! Like much research, however, it seems clear that work on the origins of communication will likely be coopted into the practical world and we may eventually see a forms of computer-computer or even human-computer interaction arising from investigations originated in the EOL debate. The final message is one of cautious optimism - we stand on the threshold of a new world but it seems that we will need to do a good deal more exploring of this new world before we find any suitable vantage points to look back upon the old. Arthur, W. S. Durlauf and D. Lane (eds). 1997. The Economy as an Evolving Complex System II. Reading, MA: Addison-Wesley Batali, J. 1994. Innate Biases and Critical Periods: Combining evolution and learning in the acquisition of syntax. Retrieved Wednesday, 9 October 2002 from http://www.isrl.uiuc.edu/~amag/langev/localcopy/pdf/batali94innateBiases.pdf Batali, J. 1998. Computational simulations of the emergence of grammar. In J. Hurford, M. Studdert-Kennedy and C. Knight (eds). Approaches to the Evolution of Language: Social and Cognitive Bases. Cambridge: CUP, 405-426. Berrah, A. 1998. Evolution Aritificielle d’une Societe d’Agents de Parole: Un modele pour l’emergence du code phonetique, Doctoral dissertation, Institut National Polytechnique de Grenoble. Bickerton, D. 1990. Language and Species. Chicago: University of Chicago Press. Bickerton, D. 2002. Foraging versus Social Intelligence in the Evolution of Protolanguage. In A. Wray (ed). The Transition to Language. Oxford: OUP, 207-226. Bonabeau, E. and G. Theraulaz. 2000. Swarm Smarts. Scientific American: March 2000: 54-61. Briscoe, E. J. 1998. Language as a Complex Adaptive System: Coevolution of Language and of the Language Acquisition Device. Retrieved Wednesday, 9 October 2002 from: http://www.isrl.uiuc.edu/~amag/langev/localcopy/pdf/briscoe98languageAs.pdf Cangelosi, A. 2001. Evolution of communication and language using signals, symbols, and words. Retrieved Wednesday, 9 October 2002 from: http://www.isrl.uiuc.edu/~amag/langev/localcopy/pdf/cangelosi01evolutionOf.pdf Cheney, D and R. Seyfarth. 1990. How Monkeys See the World. Chicago: University of Chicago Press. de Boer, B. 2000. The emergence of sound systems through self-organisation. In C. Knight, M. Studdert-Kennedy and J.R. Hurford (eds). The Evolutionary Emergence of Language. Cambridge: CUP, 177-198. de Boer, B. 2001. The Origins of Vowel Systems. Oxford: OUP. Dessalles, J-L. 2000. Language and Homonid Politics. In C. Knight, M. Studdert-Kennedy and J.R. Hurford (eds). The Evolutionary Emergence of Language. Cambridge: CUP, 62-80. Flake, Gary. 2001. The Computational Beauty of Nature. Cambridge, MA: MIT Press Glotin, H and R. Laboissiere. 1996. Emergence du code phonetique dans une societe de robots parlants. Actes de la Conderence de Rochebrune 1996: Du Collectif au Social. Paris: Ecole Nationale Superiere des Telecommunications. 113-125. Gmytrasiewicz, P. and D. Gopal. 2000. Towards Automating the Evolution of Linguistic Competence in Artificial Agents. Retrieved Wednesday, 9 October 2002 from: http://www.isrl.uiuc.edu/~amag/langev/localcopy/pdf/gmytrasiewicz00towardsAutomating.pdf Hashimoto, T and T. Ikegami. 1995. Evolution of Symbolic Grammar Systems. Retrieved Wednesday, 9 October 2002 from: http://www.isrl.uiuc.edu/~amag/langev/localcopy/pdf/hashimoto95evolutionOf.pdf Hurford, J. 1989. Biological evolution of the Saussurean sign as a component of the language acquisition device. Lingua, 77(2):187-222. Hurford, J. 1991. The Evolution of the Critical Period for Language Acquisition. Cognition, 40(3):159-201. Hurford, J. 2000. Social transmission favours linguistic generalization. In C. Knight, M. Studdert-Kennedy and J.R. Hurford (eds). The Evolutionary Emergence of Language. Cambridge: CUP, 324-352. Hurford, J.R., M. Studdert-Kennedy and C. Knight (eds). 1998. Approaches to the Evolution of Language. Cambridge: CUP. Hutchins, E. and B. Hazlehurst. 1995. How to invent a lexicon: the development of shared symbols in interaction. Retrieved Wednesday, 9 October 2002 from: http://www.isrl.uiuc.edu/~amag/langev/localcopy/pdf/hutchins95howTo.pdf Kaplan, F. 1998. A New Approach to Class Formation in Multi-Agent Simulations of Language Evolution. Retrieved Wednesday, 9 October 2002 from: http://www.isrl.uiuc.edu/~amag/langev/localcopy/pdf/kaplan98aNew.pdf Kirby, S. 1997. Competing motivations and emergence: explaining implicational hierarchies. Retrieved Wednesday, 9 October 2002 from: http://www.isrl.uiuc.edu/~amag/langev/localcopy/pdf/kirby97competingMotivations.pdf Kirby, S. 1998. Fitness and the selective adaptation of language. In J. Hurford, M. Studdert-Kennedy and C. Knight (eds). Approaches to the Evolution of Language: Social and Cognitive Bases. Cambridge: CUP, 359-383. Kirby, S. 2000. Syntax without Natural Selection: How compositionality emerges from vocabulary in a population of learners. In C. Knight, M. Studdert-Kennedy and J.R. Hurford (eds). The Evolutionary Emergence of Language. Cambridge: CUP, 303- 323. Kirby, S and J. Hurford. 1997. Learning, culture and evolution in the origin of linguistic constraints. Retrieved Wednesday, 9 October 2002 from: http://www.isrl.uiuc.edu/~amag/langev/localcopy/pdf/kirby97learningCulture.pdf Kirby, S and J. Hurford. 2001. The Emergence of Linguistic Structure: An overview of the Iterated Learning Model. Retrieved Wednesday, 9 October 2002 from: http://www.isrl.uiuc.edu/~amag/langev/localcopy/pdf/kirby01theEmergence.pdf Knight, C., M. Studdert-Kennedy and J.R. Hurford (eds). 2000. The Evolutionary Emergence of Language. Cambridge: CUP. Livingstone, D and C. Fyfe. 2000. Modelling Language-Physiology Coevolution. In C. Knight, M. Studdert-Kennedy and J.R. Hurford (eds). The Evolutionary Emergence of Language. Cambridge: CUP, 199-215. Lucas, S. 1994. Structuring Chromosomes for Context-free Grammar Evolution. Retrieved Wednesday, 9 October 2002 from: http://www.isrl.uiuc.edu/~amag/langev/localcopy/pdf/lucas94structuringChromosomes.pdf MacLennan, B and G. Burghardt. 1993. Synthetic Ethology and the Evolution of Cooperative Communication. Retrieved Wednesday, 9 October 2002 from: http://www.isrl.uiuc.edu/~amag/langev/localcopy/pdf/maclennan93syntheticEthology.pdf Mitchell, M. 1998. An Introduction to Genetic Algorithms. Cambridge, MA: MIT Press. Oliphant, M. 1996. The dilemma of Saussurean communication. Retrieved Wednesday, 9 October 2002 from: http://www.isrl.uiuc.edu/~amag/langev/localcopy/pdf/oliphant96theDilemma.pdf Oliphant, M. 1999. The learning barrier: Moving from innate to learned systems of communication. Retrieved Wednesday, 9 October 2002 from: http://www.isrl.uiuc.edu/~amag/langev/localcopy/pdf/oliphant99theLearning.pdf Pinker, S. and P. Bloom. 1990. Natural language and natural selection. Behavioral and Brain Sciences, 13(4):707-784. Power, Camilla. 2000. Secret Language Use at Female Initiation: Bounding Gossiping Communities. In C. Knight, M. Studdert-Kennedy and J.R. Hurford (eds). The Evolutionary Emergence of Language. Cambridge: CUP, 81-98. Savage-Rumbaugh, E. S. and R. Lewin. 1994. Kanzi: The Ape at the Brink of the Human Mind. New York: Wiley. Shotter, J. 1993. Cultural Politics of Everyday Life. Buckingham: Open University Press. Smith, A. D. M. 2001. Establishing Communication Systems without Explicit Meaning Transmission. Retrieved Wednesday, 9 October 2002 from: http://www.isrl.uiuc.edu/~amag/langev/localcopy/pdf/smith01establishingCommunication.pdf Steels, L. 1997. The origins of syntax in visually grounded robotic agents. Retrieved Wednesday, 9 October 2002 from: http://www.isrl.uiuc.edu/~amag/langev/localcopy/pdf/steels97theOrigins.pdf Steels, L. 1998. Synthesizing the origins of language and meaning using coevolution, self-organization and level formation. In J. Hurford, M. Studdert-Kennedy and C. Knight (eds). Approaches to the Evolution of Language: Social and Cognitive Bases. Cambridge: CUP, 384-404. Steels, L. 2000. Language as a Complex Adaptive System. Retrieved Wednesday, 9 October 2002 from: http://www.isrl.uiuc.edu/~amag/langev/localcopy/pdf/steels00languageAs.pdf Steels, L. F. Kaplan, A McIntyre and J. van Looveren. 2002. Crucial Factors in the Origins of Word-Meaning. In A. Wray (ed). The Transition to Language. Oxford: OUP, 252-271. Stonier, R and Xing, H. Y. 1995. Introduction. Complexity International:2. Retrieved Friday, 18 October 2002 from: http://www.csu.edu.au/ci/vol02/intro.html Taylor, T. 1997. Theorizing Language. Amsterdam: Pergamon. Tonkes, B and J. Wiles. 2002. Methodological Issues in Simulating the Emergence of Language. In A. Wray (ed). The Transition to Language. Oxford: OUP, 226-251. van Looveren, J. 2001. Robotic Experiments on the Emergence of a Lexicon. Retrieved Wednesday, 9 October 2002 from: http://www.isrl.uiuc.edu/~amag/langev/localcopy/pdf/vanlooveren01roboticExperiments.pdf Vogt, P. 2001. The impact of non-verbal communication on lexicon formation. Retrieved Wednesday, 9 October 2002 from: http://www.isrl.uiuc.edu/~amag/langev/localcopy/pdf/vogt01theImpact.pdf Wallman, J. 1992. Aping Language. Cambridge: CUP. Werner, G and M. Dyer. 1991. Evolution of communication in artificial organisms. In C. Langton, et al. (eds). Artificial Life II. Redwood City, CA: Addison Wray, Alison (ed). 2002. The Transition to Language. Oxford: OUP.