In the past decade, Phenomenal Intentionality Theories (PITs), which state that intentionality is ultimately grounded in phenomenal properties, have become a prominent alternative to “tracking” approaches to intentional content, such as causal or teleosemantic theories. I argue that strong versions of PIT cannot successfully account for identity between intentional contents. This is a problem because it is desirable for a theory of intentionality to be able to account for sameness and difference between the contents of intentional states.

To demonstrate that strong PITs cannot account for identity between contents, I begin by briefly laying out some of the minimal constraints of these theories when it comes to the relationship between intentional content and phenomenal character. In particular, I express this through a fairly neutral supervenience thesis which states that, when it comes to at least some basic intentional states, there can be no difference in intentional content without there being a difference in phenomenal character. An additional preliminary note is necessary for my argument: introspection plays a key role in PITs, and specifically, introspective reports about the intentional content of experience are given great importance. Mendelovici, who is one of the most prominent proponents of a strong version of PIT, is explicit in this regard, and states that our theories of intentionality must match our pretheoretical introspective judgments about the contents of our experience to be adequate. In other words, PIT should methodologically prioritize introspection over other ways of determining the content of intentional states.

After having laid out these preliminary notes, I construct a scenario that shows how even these minimal constraints, along with the methodological priority given to introspection, lead to an unacceptable tension within PIT. The scenario is as follows: a subject faces a circular gradient composed of eight continuous shades (s₁-s₈). By design, s₁ is indiscriminable from s₂; s₂ is indiscriminable from s₃; and so on. However, s₁ is easily differentiated from s₈. I provide an image of the gradient to illustrate this.

Based on this scenario, one can lead the defender of PIT into contradiction. The subject will report that the content of their experience of s₁ and s₂ is identical; that the content of their experience of s₂ and s₃ is identical; and so on. Since identity is transitive, identity between contents should be transitive too, which means that we can infer that the content of the subject’s experience of s₁ is identical to the content of their experience of s₈. However, the subject will report, based on introspection, that the content of their experience of s₁ and s₈ is not identical. We’ve stumbled into a contradiction.

The defender of PIT is faced with two unpleasant options here. One is to claim that one of the premises of the argument is false, that is, that one of the reports of the subject is false. However, this can only be done by abandoning the aforementioned priority of introspection, since all premises are supported by introspective reports. This is unacceptable because the prioritization of introspection is one of the fundamental ideas behind the program of phenomenal intentionality. The other option is to claim that the transitivity of identity does not hold for intentional contents, which is akin to saying that there can be no true identity between intentional contents, since transitivity is one of the defining characteristics of identity relations. This is unacceptable because the capacity to accommodate and explain sameness and difference of contents is a basic desideratum for any theory of intentionality.

I conclude this work by anticipating a possible counterargument. One might try to defend strong PITs by saying that there could be other factors determining content, and that PITs need not give total priority to introspection; for example, Mendelovici argues that our pretheoretical intuitions about content are sometimes related to the psychological role an intentional state plays (2018, pp. 27-8). I construct an argument to demonstrate why this is not a plausible way out for PIT. My argument has an analogous form to Mendelovici’s “mismatch argument,” which she employs to argue that tracking theories of intentional content are inadequate (2018, pp. 33-44) – it demonstrates that there could be a mismatch between the content determined by a state’s psychological role and the content determined by a subject’s introspective access to that same state. If such a mismatch occurs, I argue, the defender of PIT is constrained to give priority to what introspection dictates is the content of the intentional state. The argument can be generalized to target other kinds of potential theoretical “accessories” to PIT which would allow for alternative predictions about the content of intentional states. Defenders of PIT, then, must prioritize introspection over alternative ways of determining the content of intentional states.

The upshot of this work is that strong versions of PIT, such as the ones developed in Mendelovici (2018) and Graham, Horgan and Tienson (2007), are committed to methodologically prioritizing introspection in a way that leads to them being unable to account for identity between contents in a satisfactory manner. I suspect this is a consequence of a deeper and more fundamental problem with strong PITs, which is that “flattening” representational properties onto phenomenal properties threatens to strip intentional states of their normative nature. In other words, strong versions of PIT such as Mendelovici’s do not seem to leave space for intentional states to misrepresent; and there can be no representation without the possibility of misrepresentation. I believe more work needs to be done to investigate whether PITs can be modified to accommodate this basic feature of representation.

Bibliography:

Burge, Tyler (2007). Foundations of mind. New York: Oxford University Press.

Brewer, Bill (2011). Perception and Its Objects. Oxford, GB: Oxford University Press.

Campbell, John. (2002). Reference and Consciousness. Oxford: Oxford University Press.

Chalmers, David J. (2004). The representational character of experience. In Brian Leiter (ed.), The Future for Philosophy. Oxford University Press. pp. 153--181.

Dretske, Fred. (1995). Naturalizing the Mind. Cambridge: MIT Press.

Graham, George; Horgan, Terence E. & Tienson, John L. (2007). Consciousness and intentionality. In Max Velmans & Susan Schneider (eds.), The Blackwell Companion to Consciousness. Blackwell. pp. 468--484.

Horgan, Terence & Tienson, John (2002). The Intentionality of Phenomenology and the Phenomenology of Intentionality. In David J. Chalmers (ed.), Philosophy of Mind: Classical and Contemporary Readings. Oup Usa. pp. 520--533.

Mendelovici, Angela. (2018). The Phenomenal Basis of Intentionality. New York, USA: Oxford University Press.

Siewert, Charles P. (1998). The Significance of Consciousness. Princeton University Press.

Tye, Michael. (2000). Consciousness, Color, and Content. Cambridge: MIT Press.

Williamson, Timothy. (1994). Vagueness. New York: Routledge.

Yli-Vakkuri, Juhani, and John Hawthorne. (2018). Narrow Content, Oxford: Oxford University Press.

Representationalism about phenomenal consciousness is the view that phenomenal character is determined by representational (i.e., intentional) content. According to pure representationalism, phenomenal character is wholly determined by intentional content, whereas according to impure representationalism, it is determined by intentional content plus something else, usually some functional-cognitive property. Most representationalists endorse the impure version, plausibly because there are non-phenomenal representations and it isn't easy to distinguish in representational terms between them and phenomenal representations. But there are exceptions. Thus, according to Bourget (2010), phenomenal character is nothing but content – it is un-derived (otherwise put – original) content. Bourget's view appears to account successfully for the distinction between phenomenal and non-phenomenal representations, since (or so most philosophers believe) non-phenomenal representations do not possess their contents originally, but only derivatively: their contents are derived from the contents of phenomenal states with which they maintain some connections. (Another proponent of pure representationalism is Thau 2002. For impure representationalism see, e.g., Tye 1995, Dretske 1995, and Jackson 2004.)

The main purpose of this talk is to undermine the view that phenomenal consciousness is determined by original content. My claim is that either this view fails to distinguish between phenomenal and non-phenomenal representations, or it involves a non-representational element, and so is in fact an impure representationalist view. A subsidiary goal of the paper is to draw a distinction between two kinds of phenomenal representationalism on a dimension different from that of pure versus impure distinction – a subjectivist kind and a deflationist kind. This distinction is important for its own sake, but it will also play a role in the suggested case against Bourget's view.

The issue of original versus derived intentionality concerns the source of the contents of entities, rather than the identity of these contents. A conscious mental state (which is assumed to have its content originally) and a (mental or non-mental) non-conscious state or act (which is assumed to have its content derivatively) can share content: to ascribe to them the content that p originally and derivatively, respectively, is to ascribe to them not only different properties but also the same property, namely the same intentional content. So, I will argue, if it is the content of a conscious mental state that is supposed to fully determine its phenomenal character on the view under consideration, then according to this view the conscious state and the non-conscious one must have the same phenomenal character and (a fortiori) the non-conscious state comes out conscious. It might be thought that this predicament can be avoided in a reductive framework of intentionality. In such a framework, the determiner of a state's possession of original content is different from the determiner of another state's possession of the semantically identical derived content, and we can take the phenomenality of a conscious state to be constituted by the determiner of this state's possession of its content rather than by its content itself. This emendation of representationalism seems to reserve its spirit, and it well coheres with an important motivation for it, which is to ultimately ground phenomenality in the "natural". However, I will argue that even if we can motivate this view as a theory of phenomenality, what enables it to distinguish between phenomenally conscious representational states and non-conscious ones is precisely the fact that the determiner of the possession of original content according to it is a determiner of the specific way in which the bearer of original content possesses it, rather than of its mere possession of content. So this view is a version of impure reprsentationalism.

But perhaps representationalists can have a way out. We should distinguish between two sorts of views according to which intentionality determines phenomenality, one that takes phenomenality to be thin and another that takes intentionality to be thick. On the latter view, intentionality is “shot through with subjectivity”, as McGinn (1991) puts it (and, thus, thick), and its subjective nature enables it to determine the subjective nature that is associated with phenomenally conscious states. (Searle (1992) also seems to endorse such a subjectivist phenomenal representationalism.) On the former view, intentionality is laden with nothing but its semantic content, understood as sheer aboutness, and so phenomenality consists of nothing more (and is, then, thin). (Pure) representationalism thus understood may be said to embody a deflationist view of phenomenality, one which takes subjectivity out of consciousness or downplays it. Most current representationalist views do not take intentionality to be "shot with subjectivity", and so may be said to be deflationists. But it might seem that "original-content- representationalism" such as Bourget's can incorporate the subjectivist version and thus single out phenomenal representations. The idea might be that original content and only original content is shot with subjectivity, and it thus endows representations that possess it with phenomenality. However, we shall see that accounts of phenomenal consciousness in terms of intentionality that is shot with subjectivity are doomed to fail on pain of circularity – their explanations of what makes something phenomenal are virtus dormitiva explanations. Trivially, original-content-representationalism inherits this flaw.