Stained glass as a model for consciousness

Abstract

Contemporary phenomenal externalists are motivated to a large extent by the transparency of experience and by the related doctrine of representationalism. On their own, however, transparency and representationalism do not suffice to establish externalism. Hence we should hesitate to dismiss phenomenal internalism, a view shared by many generations of competent philosophers. Rather, we should keep both our options open, internalism and externalism. It is hard, however, to see how to keep open the internalist option, for although transparency and representationalism have not yet definitively established externalism, they have indeed made it quite intuitive. Internalism, by comparison, comes across at first sight as antiquated and ridden with difficulties. This is why I propose the Stained Glass model of consciousness. I do so with the following two aims: first, to make internalism intuitive in the age of transparency, and second, to show how to resist the many recent anti-internalist arguments. In particular, I argue that phenomenal internalism need not be epistemically worrisome, that it is compatible at once with transparency, representationalism, and content externalism, and that although it requires an error theory, this error theory is a harmless one.

Keywords:

• transparency of experience
• phenomenal externalism
• internalism
• phenomenal consciousness
• representationalism

Acknowledgements

I am grateful to Robert Van Gulick for important feedback on several aspects of this article, to Neil Mehta for a number of insightful observations, to Amy Kind, who commented on the article at the Eastern APA in 2012, and to André Gallois and Clyde Hardin, who have read early drafts.

Notes on contributor

Mihnea D. I. Capraru is a dissertation fellow within the Humanities Center at Syracuse University, where he works in philosophy of language, philosophy of mind, and philosophy of biology.

Notes

1. On another formulation, phenomenal internalism is the view that phenomenal properties supervene on the intrinsic physical properties of the perceiver. Although this overlaps largely with my formulation, it is not entirely equivalent to it.

2. Among the objects we call external, let us include somatic objects to accommodate bodily phenomenal properties.

3. Philosophers often distinguish strong representationalism from weak (Tye 2002a, 137; Byrne and Tye 2006, 241, 254; Kind 2007, 405–406; Kind 2010, 904). Weak representationalism claims merely that phenomenal properties supervene on the contents of our representations. In this article, by ‘representationalism’, I mean strong representationalism.

4. In 2009, Molyneux has proposed another thought experiment intended to model the transparency of experience. Molyneux has done so to defend a theory that I think suffers significant flaws. This, however, is not the place to criticize the theory; rather, let me criticize the thought experiment and let me show why we are better off with Stained Glass. Molyneux's thought experiment features a character named Bob. Bob is fitted with imperceptible ‘televisual contact lenses’ (Molyneux 2009, 123). These TV contacts are not quite lenses and they are not transparent. Instead, they film external reality with small cameras and they display pictures of external reality on their inner surfaces. So it comes that Bob, unbeknownst to him, perceives external reality indirectly (Molyneux 2009, 123). Molyneux maintains that TV contacts effectively model the transparency of experience. Let us, however, consider a Stained Glass alternative, one that I will argue is preferable. Consider stained contacts. Stained contacts work just like our transparent stained glass dome, except that they nest imperceptibly in Bob's eyes. Compared with TV contacts, stained contacts have multiple advantages: First, since we are trying to model transparency, it helps that unlike TV contacts, stained contacts are genuinely transparent. Second, with transparent stained contacts, it is obvious that Bob can perceive external objects rather than merely his own contacts; with opaque TV contacts, this is not obvious (notwithstanding Molyneux's arguments in 2009, 124–125). Third, with stained contacts, it is obvious that Bob is experiencing phenomenal properties as properties of external objects, whereas with TV contacts, it is not: with TV contacts, it might be that Bob experiences phenomenal properties as properties of the TV pictures and mistakes the TV pictures for regular external objects. And finally, with stained contacts, Bob perceives external objects directly; with TV contacts, however, he perceives them indirectly in a strong sense that seems to commit Molyneux to something similar to sense data (Molyneux 2009, 123).

5. For this premise, notice that even weak representationalism will serve.

6. We can distinguish shades of content externalism along at least two dimensions: First, how many – and which – representations have their contents determined extrinsically, and second, whether contents are determined entirely or only in part extrinsically. To mount the argument from content externalism, phenomenal externalists need to assume at least this much: that all representations with phenomenal content have their content determined extrinsically, and that the phenomenal component of their content is determined extrinsically at least in part.

7. It can be tricky to formulate content externalism rigorously; my formulation draws from Van Gulick (2004, 258–261).

8. To accommodate the extended mind hypothesis, we might add: or inside the extended mind, yet outside the organism.

9. To be sure, proponents of Russell's view do not always intend it as a form of physicalism; indeed, Russell himself intended it as a form of neutral monism. Nevertheless, authors such as Montero (2010) or Pereboom (2011) have recently proposed physicalistic versions of Russellian monism.