Go here for an excerpt of Daniel Dor’s book.
Linguist Nicholas Evans predicts that Daniel Dor’s The Instruction of Imagination: Language as a Social Communication Technology “will change the face of the discipline—not just of linguistics, but of the language sciences more generally.” Evolutionary humanist Brian Boyd begins a forthcoming review: “If you want to think about language and evolution, about language and experience, about language and almost anything, or about almost anything in language, then start, or start all over again, with Daniel Dor’s The Instruction of Imagination.”
Why do these experts find Dor’s account of language and humans as language-users so provocative and stimulating? TVOL is proud to feature a conversation between Boyd and Dor, along with an excerpt of Dor’s book, to explore a long overdue revolution in our understanding of language.
Brian Boyd: Language has always seemed central to what makes humans different. You summarize your evolutionary argument: “First we invented language. Then language changed us.” That’s one tight nutshell for a theory with such wide and deep implications. Can you open it up a little more?
Daniel Dor: Yes, the evolutionary argument is based on the model I developed together with evolutionary biologist Eva Jablonka, and it basically says three things. First, human language is a technological invention – a collective technological invention. Ancient human communities (as early as half a million years ago) invented the first prototypes of language, and later generations continued to develop and (sometimes) re-invent them. In this sense, there is no difference between language and other collective inventions, from the first stone tools to the Internet. Second, the original invention was made both possible and necessary by complex dynamics at the collective level, before individuals were language-ready in the cognitive (and emotional) sense: participation in the collective effort of language was made possible (to various degrees in different individuals) by cognitive and emotional plasticity. Third, as the new technology began to revolutionize human life, language began to function as a selective environment for individuals. Eventually, Homo sapiens emerged as a language-ready species. Our language-ready brains and physiologies (which are still as variable as our ancestors’) were forced into existence by language, not the other way around.
BB: You say that the technology of language is dedicated to the instruction of imagination. What do you mean by that?
DD: Well, the claim is that the uniqueness of language lies in this very specific functional strategy. All the other systems of intentional communication, used by humans and the other species that we think we understand, work with different variations of the functional strategy that I call experiential: all these systems allow for (different variations of) the communicative act of presenting: “this is my experience”. This very general characterization captures the foundational fact that experiential communication is inherently confined to the here-and-now of the communication event, where experiences can be presented. Language is the only system that allows communicators to communicate directly with their interlocutors’ imaginations, and thus break away from the here-and-now of co-experiencing: instead of presenting the experience to their interlocutors for perception, communicators translate their experiential intents into a structured code, which is then transmitted to their interlocutors and instructs them in the process of imagining the experience – instead of experiencing it. What the interlocutors do is use the code to bring back from their own memory experiences connected to the components of the code, rearrange them according to the structural configuration of the code, and construct a new, imagined experience. Language thus allows for a unique communicative act: “my experience is of this type, try to imagine”. This unique communicative strategy is the key to the enormous success of language and its influence on the human condition, and it also provides a new explanatory foundation for a very wide range of empirical issues in linguistics and the other disciplines interested in language, from psycholinguistics to literary studies.
BB: How does your theory of language differ from the assumptions dominant in and beyond linguistics ever since Chomsky first spoke out in the 1950s?
DD: We need to distinguish between Chomsky’s foundational assumption and his foundational hypothesis. The assumption was that language is essentially a uniquely human, universal cognitive capacity. The secret of language resides in the human mind. The hypothesis was that language does not just reside there: it is an innately-given property of the mind. In the last half century, the hypothesis has been thoroughly discredited in most linguistic circles: we do not carry the explicit knowledge of language in our genes. The assumption, however, that language is essentially a universal cognitive capacity, has remained intact. My theory suggests that we should finally allow ourselves to leave the assumption behind, and begin to re-think language as a social entity, a communication technology, that resides between speakers, not simply in them. Just like the social media on the Internet today, whose logic cannot be reduced to the computational activity within the personal computer, language needs to be understood as a property of the social web. Once we agree to make this move, we find that this new characterization of language reflects back on the cognitive questions, and suggests a much more reasonable account of the capacities involved with language – their properties, usage, and evolution.
BB: What difference does it make to see language as a technology we invented socially rather than a cognitive capacity evolution built up in us individually?
DD: Well, it makes a huge difference for the entire set of questions we have to ask about language. The logic of the architecture and usage of language remains a mystery as long we think about it as a reflection (of this type or the other) of the properties of the individual mind. It is revealed when we think about it in technological terms, as a logic of technological design that is constantly engaged in the effort to enhance the efficiency of imagination-instruction. The complex patterns of difference and similarity between the languages of the world are very hard to explain in cognitive terms but make simple sense if we think about them as variations in types of technological solutions. The influence of language on the way we think and experience is more similar to the influence of other technologies than to anything else. And, of course, there is simply no way to explain the evolutionary emergence of such a cognitive capacity – for two complementary reasons: if we assume that the capacity is sui generis, and that it emerged as such, we have to leave its emergence in the realm of mystery (which is exactly Chomsky’s position); if we assume that the capacity emerged from more ancient capacities that we share with the apes, we cannot explain why they never invented language, not even a much simpler one. My approach re-positions the emergence of language within the context of collective, technological innovation: language is one technological innovation in the series of innovations that made us who we are, not the first and definitely not the last. The capacity that made language possible is the social capacity of collective innovation, which is exactly what the apes lack: they understand the world and the others around them much better than previously assumed, they are very good learners and they often invent. They do not, however, invent together. For me, this is the key.
BB: In your theory, behavior drives evolution. How do you see this at work more generally?
DD: This is exactly what we’ve learned in the last two decades from evolutionary developmental biology (evo-devo). The argument is explicated most clearly in Jablonka and Lamb’s Evolution in Four Dimensions and in West-Eberhard’s Developmental Plasticity and Evolution. When their living conditions change, organisms do not just sit and wait for a random genetic mutation to help them survive. They launch processes of behavioral exploration, using their plasticity to search for solutions outside their arsenal of regular behaviors. Then there are many Ifs: If they are lucky enough to stumble upon a solution; and if they manage to stabilize it; and if later generations manage to learn the stabilized behaviour; and if the pressure to sustain the behaviour and develop it continues for a good amount of time, then eventually the behaviour will be (partially) genetically accommodated. As West-Eberhard puts it: “genes are followers in evolution”. This is exactly what allows us to reverse the order of things with respect to language evolution: the collective invention of the first prototypes of language launched a never-ending process of behavioural explorations, always based on plasticity, which on the one hand constantly changed language, and on the other hand gradually turned us into language-ready organisms.
BB: How and why do you propose our forebears invented language?
DD: The key lies in everything that they managed to achieve before language, and thus without language, in the period between around 2 million years ago to about a half a million years ago – at the levels of sociality, technology, and communication. In the simplest terms possible, individual human survival in this period gradually came to depend more and more on the others – on the overall capacity of the community to work together, invent together, fight together and so on. Communities grew in size and internal complexity; new and unique patterns of division of labor appeared; the nature of social relationships changed. All these allowed for the stabilization of a series of technological and behavioral revolutions – cooking, tool-manufacturing, big game hunting, and much more. The forms of experiential communication these ancient humans inherited from the apes were no longer sufficient to maintain the fabric of these collectivizing communities and to transfer their practical-technological knowledge from one generation to the next. A new and uniquely human form of experiential communication emerged (again, through behavioural exploration) – mimetic communication: the combination of intentional pointing, eye contact, gesture, mimicry, bodily movement, facial expression, tone of voice, song and dance that we still find much more efficient than language in the teaching of practical knowledge and the negotiation of social relationships. As Merlin Donald shows in his classic Origins of the Modern Mind, mimetic communication is the missing link between ape communication and human language.
Mimesis did much more than allow these ancient humans to communicate more efficiently. It allowed them to begin to systematically compare their experiences of the world, to learn from each other and teach each other, and construct a collective worldview. This collective capacity for what I call experiential mutual identification was the foundation upon which language emerged. As revolutionary as it was, however, mimesis was still a form of experiential, presentational communication, almost totally confined to the here-and-now of the communication event. At a certain point, this was no longer enough. Communities came to depend on experience sharing to such an extent, that they needed to find ways to share experiences that could not be directly presented to the senses – to go beyond “look, there is something there” to “there is something there, where you cannot see”. What was required was a radical change of attitude: the will and ability to trust the communicator and replace direct experiencing with imagining. This was the crossing of the Rubicon. What the inventors of language began to experiment with was thus not the technology of language, but the usage of the old tools for the new communicative function of imagination-instruction. From that moment, the constant need to raise the levels of success in instances of imagination-instruction pushed a process in which language gradually separated itself from the old tools and developed its constitutive architecture, and speakers gradually developed their language-ready minds and bodies.
BB: Linguistics has often discussed language in relation to thought. You discuss it in relation to experience. Can you explain that? What difference does it make?
DD: The conception of language as related to thought, actually to rational thought, is something we inherited from the religious and philosophical traditions involved with language – and I believe it needs to be abandoned. On the one hand, we know today that many other types of animals, who have nothing like language, are perfectly capable of rational thought. We have seen enough demonstrations of apes, parrots, and crows solving seriously complex problems to be convinced of that. We do not think in words (this is not to deny that language helps us think, and sometimes influences the way we think). On the other hand, language is used for so much more than the communication of rational thoughts, that there seems to be no reason to connect it so tightly to it. Rational thinking is one way to handle experiential complexity, one out of many, and it is entangled with everything else in experiencing. The properties of language (including its emergence) are much better explained if we take it to be a technology that takes pieces of experience as its input and produces instructions for imagination as output.
BB: You stress the “experiential gap.” What do you mean by it?
It is a foundational fact about the process of experiencing that it is private. We experience within ourselves, and our experiential dynamics are inaccessible to the others. Because we experience privately, our experiential worlds are also significantly different from each other. Even when we experience together, we do not experience the shared experience in exactly the same way. We are forever separated by experiential gaps. This reflects on our understanding of intentional communication in general: every act of intentional communication is an attempt to get something across to the other side of the gap. In experiential communication, the gap is momentarily reduced (only partially) by shared experiencing, and whatever can be shared can be communicated. Language is the only system that actually bridges the experiential gap, allowing interlocutors to imagine, within their own experiential worlds, experiences that could not be shared by the speakers. The communicative challenge here is enormous: to instruct the interlocutors in a process of imagining that would eventually result in something that is similar enough to the speaker’s intent. This is why language is such a complex and sophisticated technology, and even with all the sophistication, the levels of success it offers are not very impressive: as the literature on conversational analysis show very clearly, we spend much of our conversations repairing misunderstandings. Language is inherently fragile: the experiential gap is always still there.
BB: What’s the effect on us as social animals of being able to instruct other imaginations across experiential gaps?
DD: To begin with, the instruction of imagination across the gap opens up revolutionary venues for communication. It allows us to tell stories about past and fictive events and discuss predictions and plans for the future, give advice and make promises, exchange information about absent people (see Robin Dunbar’s account of gossip), create alliances and construct new forms of collective identity, compare our experiential worlds and negotiate a common worldview, co-operate in practical tasks and innovate together, and also manipulate other people in an unprecedented way: lying to the imagination of the other is the most powerful tool of coercion ever invented.
All this creates a uniquely human reality, in which social life takes place on two levels instead of one – the level of experience and the level of imagination. All human communities are both experiential and imagined (not just Benedict Anderson’s nation-states), and the dialectic relationship between the two levels determines much of the unique nature of the human community. This, in turn, creates a parallel, dialectic relationship within ourselves: as opposed to all the other experiencing creatures, who develop within them an experience-based worldview, we are, so to speak, of two minds. Alongside the experiential worldview that we develop privately (based on non-linguistic experiencing, alone or together), we also develop an imaginary worldview – based on everything we have been told about. The experiential worldview still follows the ancient logic of private experiencing. The imaginary worldview follows the socially-determined logic of language and discourse. Where the two worldviews correlate, all is well. They differ, however, more often than not, actually creating an internal gap within us between two separate worlds of meaning. Much of the trouble that we experience in our lives results from this internal gap.
BB: Can you pick one implication of your theory of language that makes a concrete difference in some domain of linguistics that non-linguists can understand?
DD: Let me try to show something from the domain of lexical semantics. The major question here is surprisingly difficult: how do words have meaning? So consider the word grandmother. At first sight, this seems easy: “a grandmother is a mother of a parent”. The fact of the matter, however, is that when speakers are asked to identify appropriate referents for the word, they very often turn their attention to the prototypical properties of grandmothers: nice old ladies with candy in their bag and so on. Should we then add these properties to the definition? We could, but this wouldn’t solve the problem: on the one hand, speakers feel free to use the word to refer to mothers of parents even if they are young and unfriendly; on the other, they use the word to refer to nice old ladies even if they have no grandchildren. Should we then assume that there are two separate but related words, grandmother1 and grandmother2? This is definitely not a welcome result, but how can we avoid it? This is an old and persistent problem (pertaining of course to the entire lexicon, not just to grandmother).
According to my theory, as I have already noted, we live in two worlds of meaning at the same time, the private-experiential, and the linguistically-constructed social-imaginary. Technical details aside, my claim is that the original definition of grandmother, “a mother of a parent”, resides at the linguistic level. This is what the word means as a discrete instructor of imagination: it was invented and socially accepted as a tool for the instruction to imagine mothers of parents. Which experiential memories, then, do the users of the word bring back in order to successfully imagine a grandmother? Well, in most cases these would be memories of (different) nice old ladies. So the set of prototypical properties resides at the private-experiential level. What we have is not two different words, but one word differently represented at two interconnected levels. This does not just solve the old persistent problem and explain the patterns of usage: it shows how this complexity at the word level directly reflects on the entire relationship between us and our language.
BB: Chomsky stressed Universal Grammar as a common human cognitive “software.” Evolutionary psychology often stresses the psychic unity of humankind. But in your theory, human differences play a major role. How?
DD: This is foundational to the entire project. To begin with, evolutionary processes only know how to handle patterns of variability. If we all carry the exact same software, evolutionary theory can say very little about it, which is exactly what Chomsky has been claiming through the years: that the emergence of language (or what is essential to it) is a mystery. Second, we humans are very different from each other, much more so than individuals in other species. Take every human activity and you’ll find a very long continuum stretching from individuals who have no chance to participate, through individuals who manage to various degrees, all the way to the great geniuses who push the activity forward. This is a direct result of the collectively driven logic of the process of human evolution, the ever-growing dependency that we developed on the division of labor. Third, we are very different from each other in our language usage, acquisition, and innovation. There are obviously complex patterns of similarity, but the universal speaker, and even more importantly the universal language-acquiring child, is a myth. Major empirical issues may only receive appropriate treatment if we agree to abandon the myth. Fourth, when interlocutors see the world in very similar ways (when the experiential gap between them is narrow), they don’t need many words to communicate. The enormous complexity of language is required exactly because we try to use it when the experiential gap is wide (when the gap is too wide, of course, communication is impossible.): the development of language has always been driven by the need to bridge differences. None of this implies that there’s nothing to say about the psychic unity of humankind. It only means that we have to see this unity for what it is: a socially-constructed web of countless threads of individual difference.
BB: What is there about human language that is universal? How deep does diversity go?
DD: Well, on the one hand, all languages are technologies for the instruction of imagination, and this implies a long list of universal properties. All languages are all socially constructed; they all share the same overall architecture (a symbolic landscape comprising of imagination-instructing signs, and a communication protocol that governs the actual processes of production and comprehension); their words are represented by their speakers on the two different levels we discussed above; and so on. On the other hand, languages are as different from each other as they possibly can in terms of the specific engineering solutions they employ to meet the challenges of the instruction of imagination. All languages instruct the imagination, and each language does it in its own unique way.
There’s an interesting move here: Chomsky has taken the engineering solutions (for him they were solutions to problems of cognitive representation) to be the essence of language. He thus predicted that we should find a core set of universal solutions at the bottom of language. As the evidence accumulated in the last three decades in linguistic typology shows very clearly, this is yet another myth (see Nick Evans and Stephen Levinson’s The Myth of Language Universals). There is very little at this level that is shared by all languages: what you find, again, are complex patterns of similarity and difference. So if you remain there, you lose sight of the obvious fact that language is a universal human phenomenon. My theory re-positions the fact of universality at the social-functional level and thus frees the analysis of the variable engineering solutions adopted by different languages from an unwarranted assumption.
BB: In your book, you state: “Language allows us to imagine—the freest of all cognitive processes—but it only allows us to do that on the basis of social consensus. It is based on trust, but this only makes it the most dangerous tool of deception ever invented.” What implications might your theory have for understanding politics? Can you give examples from your own political work? Is your tragic sense of Israeli-Palestinian relations at play in this quote?
DD: Much of politics (though certainly not all of it) is done by speaking (and writing). The understanding of the foundational function of language, and how it works, deepens our understanding of political discourse, collective identity, the social construction of the self, information dissemination, influence and propaganda, negotiation and compromise. It explains why real dialogue (political and other) is at the same time the most valuable tool we have and the most frustrating activity we ever invented. It re-draws the lines between the usage and language and its abuse. In this sense, yes, my own engagement in the politics of the Israeli-Palestinian conflict has tragically provided me with the best laboratory possible. Political leaders (not just in this region of the world) consistently work to widen the experiential gap between “us” and “them”. Working against that, trying to narrow the gap enough for the two sides to begin to talk, is a Sisyphean task. At the moment, and quite globally, we seem to be losing ground.
BB: What empirical work would you like to see tested or followed up from your theory?
DD: To begin with, I believe there’s already an enormous amount of excellent empirical work that is being done out of theoretical context, so to speak, and badly needs to be incorporated into a unified theory in order to show its significance. The fascinating empirical results accumulated in the study of the Sapir-Whorf hypothesis (the hypothesis that language influences the way we think and experience) is a good example. My theory positions these results at the very centre of the drama, interprets them in a new way, and thus raises an entire set of new questions that require empirical research: what is the correlation between the individual’s susceptibility to the influence of language and his or her capacity and will to engage in linguistic communication at the expense of more experiential activities? How does social identity participate in determining the extent of influence? How do different cultures, with their variable attitudes to experience and imagination, differ from each other in terms of language’s symbolic power? The same is true of many other subfields of the linguistic sciences, and I have already mentioned quite a few in this interview.
BB: You offer major challenges to central assumptions linguistics has made for over half a century. Does this provoke resistance or incomprehension? How have linguists responded to your theory, or is it too early to tell?
DD: It’s way too early to tell.