A theory of conceptual development must specify the innate representational primitives, must characterize the ways in which the initial state differs from the adult state, and must characterize the processes through which one is transformed into the other. The Origin of Concepts (henceforth TOOC) defends three theses. With respect to the initial state, the innate stock of primitives is not limited to sensory, perceptual, or sensorimotor representations; rather, there are also innate conceptual representations. With respect to developmental change, conceptual development consists of episodes of qualitative change, resulting in systems of representation that are more powerful than, and sometimes incommensurable with, those from which they are built. With respect to a learning mechanism that achieves conceptual discontinuity, I offer Quinian bootstrapping. TOOC concludes with a discussion of how an understanding of conceptual development constrains a theory of concepts.

The thesis of our commentary is that the framework used to address what are taken by Carey to be the open issues is highly problematic. The presumed necessity of an innate stock of representational primitives fails to account for the emergence of representation out of a nonrepresentational base. This failure manifests itself in problematic ways throughout Carey's book.

Carey's claim that no object representations are perceptual rests on a faulty view of perception. To delineate origins of post-perceptual (“conceptual” or “core cognitive”) representation, we need a more accurate view of perceptual representation.

It is consistent with the evidence in The Origin of Concepts to conjecture that infants' causal representations, like their numerical representations, are not continuous with adults', so that bootstrapping is needed in both cases.

Carey speculates that the representations of core cognition are entirely iconic. However, this idea is undercut by her contention that core cognition includes concepts such as object and agency, which are employed in thought as predicates. If Carey had taken on board her claim that core cognition is iconic, very different hypotheses might have come into view.

This article defends a continuity position. Infants can abstract numerosity and young preschool children do respond appropriately to tasks that tap their ability to use a count and cardinal value and/or arithmetic principles. Active use of a nonverbal domain of arithmetic serves to enable the child to find relevant data to build knowledge about the language and use rules of numerosity and quantity.

Carey proposes that the acquisition of the natural numbers relies on the interaction between language and analogical processes: specifically, on an analogical mapping from ordinal linguistic structure to ordinal conceptual structure. We suggest that this analogy in fact requires several steps. Further, we propose that additional analogical processes enter into the acquisition of number.

We need not propose, as Carey does, a radical discontinuity between core cognition, which is responsible for abstract structure, and language and “Quinian bootstrapping,” which are responsible for learning and conceptual change. From a probabilistic models view, conceptual structure and learning reflect the same principles, and they are both in place from the beginning.

Carey focuses her theory on initial knowledge and Quinian bootstrapping. We reflect on developmental mechanisms, which can operate in between. Whereas most of the research aims at delimitating early cognitive mechanisms, we point at the need for studying their integration and mutual bootstrapping. We illustrate this call by referring to a current debate on infants' use of featural representations.

Placeholders enable conceptual change only if presumed to be relevant (e.g., lead to the formation of true beliefs) even though their meaning is not yet fully understood and their cognitive function not yet specified. Humans are predisposed to make such presumptions in a communicative context. Specifying the role of the presumption of relevance in conceptual change would provide a more comprehensive account of Quinian bootstrapping.

In order to defend her claim that the concept object is biologically determined, Carey must answer Quine's gavagai argument, which purports to show that mastery of any concept with determinate reference presupposes a substantial repertoire of logical concepts. I maintain that the gavagai argument withstands the experimental data that Carey provides, but that it yields to an a priori argument.

In this book, Carey gives cognitive science a detailed account of the origins of concepts and an explanation of how origins stories are essential to understanding what concepts are and how we use them. At the same time, this book's details help highlight the challenge of explaining how conceptual change works with real-world concepts that often have heavily degraded internal content.

In this commentary I argue that the notion of incommensurability can be extended to the child's developing theories of mind. I use Carey's concept of Quinian bootstrapping and show that this learning process can account for the acquisition of the semantics of mental terms. I suggest a distinction among three stages of acquisition and adopt the theory–theory of conceptual development.

Carey argues that the aspects of categorization that are diagnostic of deep conceptual structure and, by extension, narrow conceptual content, must be distinguished from those aspects that are incidental to categorization tasks. For natural kind concepts, discriminating between these two types of processes is complicated by the role of explanatory stance and the causal history of features in determining category structure.

We dispute Carey's assumption that distinct core cognitive processes employ domain-specific input analyzers to construct proprietary representations. We give reasons to believe that conceptual systems co-opt core components for new domains. Domain boundaries, as well as boundaries between perceptual–motor and conceptual cognitive resources may be useful abstractions, but do not appear to reflect constraints respected by brains and cognitive systems.

Carey holds that the study of conceptual development bears on the theories of concepts developed by philosophers and psychologists. In this commentary, I scrutinize her claims about the significance of the study of conceptual development.

It's unclear what narrow content is interpersonally shared for concepts that don't originate from core cognition yet are still definitionally and interpretationally primitive. A primary concern is that for these concepts, one cannot draw a principled distinction between inferences that are content determining and those that aren't. The lack of a principled distinction imperils an account of interpersonally shared concepts.

There must be innate conceptual machinery, but perhaps not as much as Carey proposes. A single mechanism of Perceptual Meaning Analysis that simplifies spatiotemporal information into a small number of conceptual primitives may suffice. This approach avoids the complexities and ambiguities of interactions between separate dedicated analyzers and central concepts that Carey posits, giving learning a somewhat larger role in early concept formation.

Carey rightly rejects the building blocks model of concept acquisition on the grounds that new primitive concepts can be learned via the process of bootstrapping. But new primitives can be learned by other acquisition processes that do not involve bootstrapping, and bootstrapping itself is not a unitary process. Nonetheless, the processes associated with bootstrapping provide important insights into conceptual change.

In order to develop sophisticated models of the core domains of knowledge that support complex cognitive processing in infants and children, developmental psychologists have mapped out the content of these knowledge domains. This research strategy may provide a blueprint for advancing research on adult cognitive processing. I illustrate this suggestion with examples from analogical reasoning and decision making.