Only human beings have a rich conceptual repertoire with concepts like tort, entropy, Abelian group, mannerism, icon and deconstruction. How have humans constructed these concepts? And once they have been constructed by adults, how do children acquire them? While primarily focusing on the second question, in The Origin of Concepts , Susan Carey shows that the answers to both overlap substantially.
Carey begins by characterizing the innate starting point for conceptual development, namely systems of core cognition. Representations of core cognition are the output of dedicated input analyzers, as with perceptual representations, but these core representations differ from perceptual representations in having more abstract contents and richer functional roles. Carey argues that the key to understanding cognitive development lies in recognizing conceptual discontinuities in which new representational systems emerge that have more expressive power than core cognition and are also incommensurate with core cognition and other earlier representational systems. Finally, Carey fleshes out Quinian bootstrapping, a learning mechanism that has been repeatedly sketched in the literature on the history and philosophy of science. She demonstrates that Quinian bootstrapping is a major mechanism in the construction of new representational resources over the course of childrens cognitive development.
Carey shows how developmental cognitive science resolves aspects of long-standing philosophical debates about the existence, nature, content, and format of innate knowledge. She also shows that understanding the processes of conceptual development in children illuminates the historical process by which concepts are constructed, and transforms the way we think about philosophical problems about the nature of concepts and the relations between language and thought.
It was exciting and thought-provoking to read this as a philosopher. Carey is a developmental psychologist, and she grapples with directly philosophical questions from her scientific angle; and moreover, she is clearly well-versed and sensitive to the history of philosophy that bears on these questions. These questions include: How do we come to acquire new concepts, and how do we distinguish between concept acquisition and mere revision of beliefs regarding the objects represented by a concept? Which concepts are innate, and how do they figure in this concept acquisition process—what constraints do they bear on how we come to learn about the world? Is mental content "narrow" or "wide," i.e., does the inferential role that a concept occupies in the mind determine the concept, or do the happenings in the world to which the concept refers determine it? I found Carey's project, moreover, deeply dissatisfying in certain ways, and reading her work helped me get clearer on what I care about. I'll first summarize her views, and then ramble a bit about my response.
First, it's useful to distinguish three ways by which a concept may be innate: First, one does not have to rely on past learning or experience in order to experience the world as involving the properties explained by these concepts. Second, this concept cannot be broken down into a complex of other concepts, definitionally. Third, this concept cannot be understood in terms of appeal to other concepts, interpretationally. Carey is primarily interested in investigating concepts that are innate in the first sense; but her work has implications for issues pertaining to innateness in the other two senses.
Carey argues, against the traditional empiricist tradition (e.g., Lock to Piaget and Quine), that there are more innate concepts than just the sensory/perceptual or sensorimotor. We don't only have innate concepts of color and shape, for example, which allows infants to experience the world as consisting of things with color and spatial properties. Instead, we also have agent-related and mathematical innate concepts. Infants, for example, from the get-go experience the world as having certain agents, who have goals and pursue them; they also experience the world as having self-same objects, which are quantifiable, and distinguishable from other qualitatively identical objects. Carey argues that these innate concepts are unshakeable. They yield conceptual schemes and intuitive theories, however, which may dramatically change over time, and be understood as 'discontinuous' with one another, in the sense of being incommensurable. These innate concepts allow us to form intuitive theories about how things work, as children, and then these theories enable us to form brand new theories, as we grow older, which replace the old ones. But these innate concepts stay the same throughout. We are incapable of having experiences or thoughts that leave out or violate these concepts.
Carey, moreover, proposes a developmental mechanism that allows these innate concepts to contribute to the production of new, more complex concepts. She is inspired by Quine's proposal of "bootstrapping," and certain theories in the history of science about how different scientific paradigms emerge. This process is a matter of an infant's relying upon innate concepts to make sense of their experiences. But the infant comes to relate to these concepts as mere "placeholders"; whatever meaning of experiences these concepts enable are registered, by the infant, as tentative and incomplete. This allows the infant to formulate new hypotheses, which go beyond or against what innate meaning is found in their experiences, but is necessarily dependent upon the constraints given by these innate concepts. When hypotheses bear out as fruitful, the infant will acquire new, more complex concepts, which again can be related to as placeholders.
Now, here are some of my responses to this work. I was dissatisfied by that at no place in the book does Carey make explicit the presumed functionalist picture of the mind, which must be true, in order for many of her claims to be coherent. For example, Carey assumes that the same concept can be manifested in either explicit or implicit forms. The concept 'shape', for example, can be implicitly manifest when we experience the world as involving objects with shapes, and then it is explicitly manifest when we reflect upon this concept and draw inferences around it. But why think that there is a single entity, a concept, that is preserved in tact across these perceptual and reflective-propositional cases? There are many explanations that could be given that do not assume this. For example, there might be a constellation of subpersonal mental processes that explain how we can perceive shape, but then when we explicitly think about the concept 'shape', various language- and knowledge-related subpersonal mental processes get newly activated, and these allow us to think about the concept 'concept' in the first place, which we then rely upon in order to think about a particular sort of concept, one that is of shape—and doing this requires that we remember our perceptual experiences, and 'shape' as it operates implicitly. In other words, across the perceptual and reflective-propositional cases, there is a constellation of subpersonal mental processes implicitly at work, and then whatever we come up with that is 'explicit' just is a wholly different thing, ontologically speaking, than those subpersonal processes.
Now, Carey's assumed view of how concepts work in compatible with only certain functionalist schemes that model the mind. On those models, 'concept' must be a self-same functional role, which can be inputted or treated more generally by other mental entities that have functional roles that capture what it means to perceive v. propositionally reflect upon something. This concept must quite directly be capable of being operated upon by either the perceptual or propositional processes. But why assume the mind is structured like this? Here's a tentative alternative: Perhaps once we use language and reflect, there are many options of ways of making sense of our perceptual experiences. There's no single explicit/propositional notion of shape that we must arrive at. There are various practical, personal, and cultural circumstances at play, which bear upon which propositional notion of shape is achieved. And this scheme would be generalized: there are many messy functionally-definable mental processes that contribute to the 'translation' from perceptually-apprehended meaning to propositionally-apprehended meaning, so that these two sorts of "meaning" (which we talk about when we talk about concepts at large), are of wholly different ontological characters, and have different functional roles for downstream mental activity.
The bigger picture concern here is that without examining these deeply implicit functionalist assumptions, we can't make real progress regarding understanding the phenomenon that gets called 'concept acquisition' or 'concept formation'.
There is a second idea that kept on surfacing in my reading experience. I've been thinking about emotion generally these days. Emotion is onto- and phylogenetically primitive. Emotion moreover is co-defined and causally co-concomitant with a certain sort of "meaning"; this fact is captured by the folk intuition that "emotion colors how we see the world." For example, a newborn infant will be capable of distress and fear; the flip side of this is that the infant is sensing certain objects in the world as bearing some threatening "meaning." Where does such emotional meaning figure into this overall project of understanding concepts, which philosophers and cognitive scientists alike have been working upon? This emotional meaning is not reducible to any of the three sorts of conceptual meaning that Carey posits as innate and fundamentally basic: the sensory, the potential of agency and holding goals of others, and mathematics. A preliminary note is that it seems that emotion drives what infants will more carefully attend to, when they deal with objects, learn about the world, and acquire these sort of "formal" concepts that Carey talks about. For example, if cold sensations typically are found whenever an infant experiences fear, then the infant will attend more closely to coldness, as an innate sensory concept. The inferences the infant will draw, or the inferential roles that this innate concept ultimately leads into or produces, will moreover be guided by this emotional meaning; the infant will be disposed to draw inferences pertaining to how coldness is bad, for example, rather than inferences pertaining to formal, non-existential or non-personal aspects of coldness.
I have the suspicion that understanding emotional meaning and its innateness can help refine the deeper functionalist assumptions that seem to structure most of philosophy and cognitive science. We can maybe arrive at a better picture of the basic parts of the mind, and how those parts interact or relate to one another, by thinking about emotion, which is so evolutionarily and developmentally primitive.
A last note: I'd highly recommend this book to anyone interested in the mind. If you're doing this for the sake of addressing philosophical issues, you only need to read chapters 1, 2, 12, and 13. There, Carey deals with the substantive claims she makes. In the other chapters, Carey explores applied levels of these claims, and these details are not useful for addressing philosophical issues.
I'm not educated well enough in this area of science to really give a readworthy review; instead, take a look at this summary.
There is a thing called CORE COGNITION. It means the part of your mind (or, to be exact, of the whole of your concept structure) that is innate; you never learned these things, you were born with them. These are what you start with; they're not concepts in the linguistic sense, but they give you the basis of the world around you. These include being able to count to maybe 5; seeing "objects" instead of just continuous sensory data (so instead of white-white-white-green-white you see a tree amidst the snow - "tree" is an object that you later conceptualise as a type of object); and so on.
There's a lot of evidence for these things. Mainly the evidence is measuring how surprised babies are in odd scenarios. If you make it seem like 1+1 is 3, babies get confused. So they probably expected 2; numbers are innate.
There's more evidence. And more.
Finally we get to ask what the ramifications are. How are our adult minds like they are? How does the learning process start off from core cognition? It seems it's mostly bootstrapping, so you learn something by not really understanding it at first but slowly concepts are formed.
Now a (rather heavy - probably the heaviest available) quote from the book:
(1) Relations among symbols are learned directly, in terms of each other; (2) symbols are initially at most only partly interpreted in terms of antecedently available concepts; (3) symbols serve as placeholders; (4) modeling processes – analogy, inductive inferences, thought experiment, limiting case analyses, abduction – are used to provide conceptual underpinnings for the placeholders; (5) these modeling processes combine and integrate separate representations from distinct domain-specific conceptual systems; and (6) these processes create explicit representations of knowledge previously embodied in constraints on the computations defined over symbols in one or more of the systems being integrated.
And that's where Carey leaves us.
It's not a bad book. It's just not for a really wide audience, if you expect to just read it and "learn something".
Perhaps the best overview of the last 30 years of research in concept acquisition. The data is often fascinating and though the book is polemically written, it is still quite judicious in its claims. Carey does not dodge any of the philosophical issues; rather, she does her best to struggle with them head on. Although I don't agree with the central bootstrapping account, it's hard not to find this book to be one of the most impressive books in the history of cognitive science.
It turns out that cognitive psychologists around the world have based their entire careers around inducing surprise in infants. Anyone aspiring to become an expert in the field should be sure to practice their baby-startling techniques, or else they could never hope to learn anything about the way the human mind works.
This book was written for people already well-versed in the field of experimental cognitive psychology. Terms are defined at random, sometimes when introduced, sometimes much later if at all, and often used in an integral manner, so it's difficult to understand what's being said if you don't have an intimate grasp of all the associated connotations and implications of the terms.
Most of 'The Origin of Concepts' consists of recounting baby-startling experiment after baby-startling experiment. The rest is sort of repetitive, dry blather that sort of dulls the brain to the point where I'm not sure if I ever really did learn where concepts originate from. Somewhere in the brain, I imagine.
Clearly I'm biased, but this book is comprehensive in content and written with extreme clarity. I still have questions about the explicit/implicit distinction, Fodor's endorsement of information semantics, and how the referential extension from concept to object in the world is causal (what is the causal chain?). The last chapter definitely needs a reread, especially regarding Fodor's thoughts on reference and natural kinds generally.
This is a very interesting if also sloppy and deeply problematic book.
The sloppiness is Carey's fault entirely: typos abound, even in sub-headings, and some of her sources do not show what she claims they do. In one case, she cited one of her own studies as suggesting the direct opposite of what it actually did. This confused me.
Many of The Origin of Concept's faults can be blamed on the field. This is a very insulated debate among a small group of people that clearly lacks diverse perspectives. As it is, Carey (and others) can jump from skepticism to naivety as it suits them. She mentions Wittgensteinian worries about projecting our conceptual systems onto/into children before launching off into an entire chapter on "how children progress like scientists." This also confused me.
The book also ends a long distance from where it begins. The initial central topic of core cognition is rather besides the point by the time Carey arrives at conceptual change, serving only to (not quite) ground her favoured learning process. But it is only near the end of the book that Carey tells us what this incredibly vague process is. So why bother with the beginning? This confused me very much indeed.
Despite these problems (and many more), The Origin of Concepts has plenty of potential outside cognitive science. Educational reform is tackled within the book itself, but I have been wondering whether the discussion on moving between incommeasurable conceptual systems couldn't be translated to phenomenology (moving between worlds) and applied to political problems as well. A book on new ways of thinking and the manner in which they are attained should be tantalizing for socialists, communists and libertarians alike.
I didn't actually read all of this book, but it's like 500 pages long and I read enough in three days to work it into my dissertation and I'm giving myself credit for that.