Review of Egan F. Deflating Mental Representation. Cambridge, MA: MIT Press, 192, 2025.Abstract
This review critically examines Frances Egan’s Deflating Mental Representation (MIT Press, 2025), a seminal work in the philosophy of mind that challenges traditional conceptions of mental representation and intentionality. Egan argues for a deflationary account wherein ascriptions of content to mental states are pragmatically grounded and serve primarily as explanatory tools, rather than reflecting substantive representational relations. The book navigates scientific, everyday, and philosophical contexts, showing how content attribution functions as a “gloss” that aids explanation without committing to metaphysical assumptions about representation. While Egan’s approach offers solutions to long-standing debates, it also faces critical tensions regarding the individuation of mental states and the role of the external environment. This review highlights both the originality and the philosophical challenges of Egan’s position, underscoring the book’s contribution to ongoing debates in cognitive science and the philosophy of mind.
Keywords:
Deflating; Mental; Representation; Frances Egan
Introduction
Frances Egan’s Deflating Mental Representation offers a cogent and sophisticated examination of one of the most enduring issues in the philosophy of mind: the nature and underpinnings of mental representation. Egan addresses the core question of how mental states come to represent entities, properties, and states of affairs distinct from themselves, ranging from familiar objects at hand to distant, abstract, or even wholly fictional entities. For instance, as I sit at my desk in Rio de Janeiro and visually register the coffee cup before me, it is equally possible for me to think about Paris, Napoleon, Alpha Centauri, the concept of justice, or the number two, entities and properties that are spatially, temporally, or conceptually remote from direct sensory experience.
This capacity leads inexorably to a central philosophical problem: how do mental states acquire their representational content? More precisely, what endows certain states of mind or brain with the ability to represent not only present objects but also objects too distant, abstract, or causally isolated to exert any immediate influence? This is the classical problem of intentionality. Contemporary theories of mental representation typically answer by positing a substantive “representation relation” linking a mental state to its intentional object, and then seek to specify the nature of this relation such that it picks out the targeted entity among myriad alternatives.
Overview of Egan’s Deflationary Alternative
Egan offers a critical evaluation of these traditional accounts and advocates an original “deflationary” alternative. She implores us to reconsider the very role of mental representation as understood across a spectrum of domains: as an explanatory construct in cognitive science, as a resource in everyday practices of behavior prediction and explanation, and as a central feature of reflective discourse about perceptual experience. In analyzing how the concept of mental representation is deployed in each of these domains, Egan arrives at a deflationary account, articulated through four central theses:
Interpretation, Not Substance: The interpretation of a mental state as a representation does not rest upon any substantive “representation relation” connecting it to what it allegedly represents, be it a coffee cup, Paris, or Napoleon. Content Is Non-Essential: Representational content is not a necessary property of mental states; it is possible for identical types of mental states to possess varying contents, or even lack content entirely. Pragmatic Grounding: Ascriptions of content to mental states are invariably motivated by pragmatic considerations. Content attribution serves to highlight, for specific explanatory purposes, features of mental activity that are not innately intentional. External Sortalism: The external world is employed as a scaffold or model to gloss mental states, a thesis Egan terms “external sortalism.”
At its broadest, the attribution of content to mental states allows us to model and render tractable processes that are otherwise private and opaque, both for scientific inquiry and everyday explanatory practice. As Egan aptly expresses, content is a “gloss” serving the specific purposes of a given explanatory context. In subsequent chapters, she investigates the nuances of what is glossed and how this varies across the different contexts in which content attributions are enacted.
Summary of the Book’s Structure and Arguments
The book’s initial chapters examine scientific approaches to the mind, particularly the cognitive revolution of the past six decades, during which mathematical and computational methods have been monumental in addressing questions of cognition. Computational models of perception, language, reasoning, and related faculties purport to explain how individuals manage sophisticated forms of environmental interaction. Within these scientific paradigms, the notion of representation is frequently invoked. In the opening chapter, Egan offers and defends a deflationary interpretation of this discourse, arguing that content attribution in the sciences of mind is irreducibly pragmatic.
By positing that content ascriptions are guided by the explanatory aims and pragmatic interests of theorists, Egan resists the ambitious philosophical project of “naturalizing” intentionality, that is, providing sufficient non-semantic or non-intentional conditions for ascribing particular content to mental states (as per “naturalism” conceived as ontological reductionism). In the second chapter, Egan contrasts ontological naturalism with what she names “naturalism as a methodological stance.” She contends that computational neuroscience and psychology can only fulfill the promise of naturalistic inquiry into the mind if they abandon essentialist commitments to meaning and intentionality. Concepts like content and intentionality cannot be simply dispensed with; rather, they are indispensable for ensuring that mathematical and mechanistic explanations retain their relevance to genuinely mental phenomena and that naturalistic sciences continue to be about the mind itself.
In chapters three and four, Egan extends her deflationary approach to belief and perceptual experience, respectively. She refutes the conception, prevalent in much of contemporary philosophy, that beliefs and perceptual experiences are states of subjects standing in fixed, substantive relations to worldly objects. On Egan’s account, such states are better understood as properties of subjects, which we model through features of the environment. Beliefs, in particular, are treated as linguistic constructs endowed with syntax, content, and truth conditions. Attributing such properties to beliefs is useful as a “gloss” or model; it does not commit us to the view that beliefs are literally linguistic. This approach facilitates the prediction, explanation, and regulation of behavior without inflating our ontological commitments.
Similarly, Egan challenges the representationalist doctrine in the philosophy of perception, which holds that differences in the phenomenal character of perceptual experiences supervene entirely on differences in representational content. Instead, she proposes “external sortalism,” influenced by traditional adverbialism. On this account, perceptual experiences are not inherently representational, though representational content may be pragmatically ascribed across various explanatory contexts. The central insight of external sortalism is that the external world provides a framework for categorizing and understanding perceptual experience; as before, representational content is a context-sensitive gloss. This approach preserves the explanatory benefits of representationalism, such as the assessment of perceptual accuracy and the rational grounding of belief, while sidestepping contentious ontological commitments to substantive representational relations.
Critical Evaluation
Collectively, Egan’s book offers a compelling and innovative perspective on the ascription of mental content in computational cognitive science, in common-sense explanations, and in our understanding of perception. Her deflationary view is not meant as a revisionist repudiation of the established roles of representation in science or everyday life. Rather, Egan intends to free both scientific and ordinary practices from a priori philosophical constraints regarding legitimate explanatory entities or properties. These practices, she contends, should be open to empirical scrutiny and may be captured by an empirically adequate deflationary account. Despite the originality and elegance of Egan’s deflationary position, several substantive challenges arise. This review will focus on what appears to be the most fundamental of these. Egan’s pragmatic stance is encapsulated in three key theses:
There is no substantive, fixed relation between the vehicle of representational content and the content itself. Representational content is not an intrinsic, essential feature of a mental state. Content ascription is always pragmatically grounded.
From these premises, it follows that the same “vehicle” could support different representational contents, depending on the pragmatic aims of researchers operating within different scientific or explanatory contexts.
However, if the real or causal connection between vehicle and content is severed, one risks lapsing into an anachronistic individualist or internalist model of mental state individuation. As Tyler Burge (1979) compellingly argued, Oscar on Earth and his counterpart on Twin-Earth do not merely represent different contents (“water” for Oscar, “twater” for Twin Oscar, per Putnam, 1975), but are in fact in distinct doxastic mental states, as they interact with different environmental properties. Burge’s anti-individualism maintains that mental states themselves change as a function of environmental context; the environment individuates not only the content, but also the mental states themselves. Egan summarizes this position as follows:
Externalists hold that physically identical subjects in different environments may have thoughts with different (‘wide’) contents. Earth-bound Oscar has water thoughts, while his counterpart on Twin-Earth, where the local oceans and rivers are filled not with H₂O but a superficially similar but chemically different liquid, has twater thoughts” (Egan, 2025, pp. 106- 107, emphasis original).
Given Egan’s rejection of externalism, a pivotal question arises: How, according to her deflationary account, are vehicles of content type-individuated? She appears to embrace a version of adverbialism (see 2025, p. 117), making it conceivable that mental-state vehicles are individuated by their intrinsic physical properties, as Papineau (2014) suggests. For example, the sentence “Paris is south of London” could be individuated by its syntactic and physical properties (font, color, size, etc.), or as Davidson’s (1968) classic phonetic example shows, a string of sounds might have identical physical structure yet differ in meaning depending on linguistic context (“Empedocles leaped” in English versus “Empedocles liebt” in German). Thus, the same vehicle may carry different contents, its representational power conferred by its use within particular linguistic or social practices (Papineau, 2021).
Alternatively, one might posit “narrow representationalism”-the claim that mental states are individuated by “narrow contents” shared by physical and functional duplicates in different environments. Egan acknowledges that the postulation of narrow contents is often motivated by the explanatory demands of behavior prediction (see 2025, p. 107). Thus, one might explain Oscar and his twin’s similar search behaviors in terms of a shared “watery-thought.” Nevertheless, this approach seems unsatisfactory: the behaviors in question are only superficially similar, as their underlying environmental relations differ. When the individuation of mental states is wide (environment-dependent), so too is the individuation of the corresponding behaviors. Oscar’s search is for H₂O, whereas his twin’s is for XYZ.
Yet, Egan appears to also abandon internalism, raising the issue anew: How are vehicles of content type- individuated under her deflationary scheme? Her proposed answer, external sortalism, assigns the external world a central role in individuating experiential type (see 2025, p. 115). Egan’s deflationary framework purports to resolve the debate over wide versus narrow content with notable elegance:
The deflationary view enables an elegant resolution of the wide vs. narrow content dispute. Recall that, on this view, contents serve to pick out the causal complexes that constitute the attitudes, but content is not to be construed as an essential property of the attitudes themselves. Wide (externally determined) content glosses on the causal complexes underlying behavior allow us to characterize these states in terms of their publicly available causes and effects, while at the same time respecting the intuition that has motivated internalists to postulate narrow content-namely, that physically identical ‘twins’ in different environments are psychologically identical. Different wide contents (water vs. twater contents) can serve to pick out the same underlying causal complex in the twins, much as different numbers-say, 212 and 100 in the Fahrenheit and Celsius scales, respectively-can be used to pick out the same underlying physical magnitude” (Egan, 2025, p. 107).
On careful inspection, however, Egan’s analogy falters. The numbers “212” (Fahrenheit) and “100” (Celsius) denote the same wide property, namely, the boiling point of water (H₂O), whereas “water” and “twater” refer to different chemical kinds (H₂O vs. XYZ). The upshot is that there is no viable intermediate position between externalism and internalism: either the vehicle of content (the mental state) is narrowly individuated (by intrinsic properties or narrow content), or it is widely individuated, in which case Egan’s deflationary approach is undermined.
Conclusion
Frances Egan’s Deflating Mental Representation is an important and innovative contribution to the philosophy of mind and cognitive science, challenging deeply entrenched assumptions about the nature and function of mental content. Although her deflationary account prompts significant philosophical questions, particularly regarding the individuation of mental states and the viability of a purely pragmatic approach to content ascription, it nevertheless provides fertile ground for further discussion and advances our understanding of the explanatory roles of content across scientific, ordinary, and phenomenological domains.
References
- Burge, T. (1979). Individualism and the mental. Midwest Studies in Philosophy, 4(1), 73-122.
- Davidson, D. (1968). On saying that. Synthese, 19(1-2), 130-146.
- Papineau, D. (2014). I-The Presidential Address: Sensory Experience and Representational Properties. Proceedings of the Aristotelian Society, 114(1 pt1), 1-33.
- Papineau, D. (2021). The Metaphysics of Sensory Experience. Oxford: Oxford University Press.
- Putnam, H. (1975). The Meaning of "Meaning". In: Mind, Language and Reality: Philosophical Papers Volume 2. Cambridge: Cambridge University Press.
