:: Article

epistemic forces and perception

Susanna Schellenberg interviewed by Richard Marshall.

Susanna Schellenberg is the badass philosophical Shosanna of the puzzles of perceptual experience and consciousness. She is forever going deep to try and make sense of what is going on and why, is an analytic with broad horizons, thinks that perception isn’t conceptually structured, doesn’t think we need sense-datum theories, explains hallucinations very differently from the orthodox approach, aims at a unified view, is Aristotlelian about types, thinks there’s a continuum between imagination and belief, is more influenced by vision science than xphi and thinks gender issues in academic philosophy will take a very long time to be sorted. Glourious!

3:AM: What made you become a philosopher? Were you always philosophizing?

Susanna Schellenberg: As a little kid I had the common obsessions about whether others see colors the way I do. But most kids have these concerns. What was more important to my becoming a philosopher is that I grew up surrounded by a lot of chaos. My Dad is a physicist by training and my Mom a cellist–though both ended up in Beirut, Lebanon and then in Islamabad, Pakistan trying to bring peace and prosperity to those regions. Early on I had the urge to try to make sense of what on earth was going on around me and why. This urge initially led me to math and maybe not quite as obviously to music. I read various philosophy books I found in the bookshelves of my parents and grandparents early on (Marx, de Beauvoir, Arendt, Plato and the like), but I doubt I understood much of what was going on. And we had a philosophy class in high school, in which, as far as I can remember, we mostly talked about love and sex. Neither experience prompted me to pursue philosophy. When I went to college, my aim was to study math. But then when I took my first logic seminar, I fell in love with philosophy. I was attracted to the questions asked and to the rigorousness in which they were tackled. I enjoy getting to the bottom of complex issues that seem central to the human condition. I have found philosophy a place where this inclination is valued.

3:AM: One of the key subjects you philosophise about is perception. You’re roots are European – is this interest perhaps initially influenced by a European heritage? I guess I’m thinking about the influence of Husserl and Heidegger on phenomenology and whether and how this influenced your philosophical interests at all?

SS: While I’m partly Swiss and partly Norwegian and so I suppose European, my philosophical education—whatever that means—has always been in the analytic tradition. That said, even in strongly analytical departments in Europe there is more of an appreciation of the history of philosophical questions than in comparable departments on this side of the Atlantic. So I may have read more Kant, Hume, and Aristotle as an undergraduate than most people who were introduced to philosophy here in the USA. Being knowledgeable about the history of philosophy is a way to broaden one’s horizon. We’re all very stuck in our time and being aware of how people thought about issues in very different ways is a great way not to be too stuck in the particular questions and conceptual frameworks of our time.

The role of perceptual experience in our cognitive lives was central to philosophical thinking throughout the history of philosophy. It was only in the early and mid 20th century that it was largely ignored for a few decades. So yes the fact that I had read these great philosophers as an undergraduate, certainly helped me get interested in perceptual experience. Awareness of the rich history of the topic also helps see the problem of perception connected to a large range of issues: how it yields knowledge of particulars in our environment, justifies beliefs, grounds language, and is instrumental to our ability to navigate the world. I was fascinated by perception since I can remember. I haven’t yet read any Heidegger. I read some Husserl only towards the end of my time in graduate school in the USA.

3:AM: One of the questions about perception is whether we can only be conscious of conceptually structured content. Perhaps before going into your response to the issue you might lay out the issue for newcomers and say why non-conceptually structured perception is thought to be such a difficulty?

SS: This debate is in part terminological. Depending on how concepts are understood it is more or less plausible to think of experience as conceptually structured. Concepts have been understood in terms of mental representations, stereotypes, functional roles, and in terms, and inferential roles to name just a few standard candidates. Nonconceptual structures have been understood in terms of image-like or map-like representations, or simply in terms of the idea that we represent naked properties and objects without representing them in terms of employing concepts.

Reasons to think that perceptual content is nonconceptually structured are to account for the fineness of grain of perceptual experience and the richness of perceptual experience. The idea is that perceptual experience is much richer and finely grained than our concepts. For example, our color concepts are much more course grained than the color shades we are able to discriminate between in perception. If that’s right (and on certain notions of concepts it is), then that’s a reason to think experience is nonconceptually structured. Another reason is that non-rational animals have perceptions, but don’t have concepts, so perceptual content cannot be conceptually structured. Whether this is a good reason depends again on what notion of concept one is operating with. After all, on certain notions of concepts it’s unproblematic to attribute concepts to non-rational animals.

One standard reasons for thinking that perceptual content is conceptually structured is that on a Fregean understanding of content it’s not clear that content could not be conceptually structured. I don’t think this is a good reason for accepting the thesis that perceptual content is conceptually structured. In a number of papers, I’ve developed a Fregean notion of perceptual content in terms of employing perceptual capacities that are not conceptual capacities. So I don’t think that a Fregean view of content must lead one to the thesis that perceptual content is conceptually structured.

A second reason for thinking that perceptual content is conceptually structured is that if singular thoughts or perceptual beliefs inherit their content from perceptual experience, then the content of experience must be structured similarly to the content of belief. I don’t think this is a good reason for accepting the thesis that perceptual content is conceptually structured either. After all, singular thoughts can be based on perception without their content being exactly like the content of the perception.

A third reason for thinking that perceptual content is conceptually structured is that perceptual experience justifies beliefs about the world and only something that is conceptually structured can justify beliefs about the world. So the content of perceptual experience must be conceptually structured. I don’t think this is a good reason either. After all, all we need for experience to play a justificatory role is that it’s content is propositionally structured. But content can be propositionally structured without being conceptually structured. Moreover, there are reasons to question that something must be propositionally structured to provide evidence. But that’s a long story.

I think the content of experience is nonconceptually structured: It ensues from employing perceptual capacities, more specifically, discriminatory, selective capacities. When I see the white cup on my desk, I employ capacities by means of which I discriminate the whiteness of the cup from the surround colors. Similarly, I employ capacities by means of which I discriminate the shape of the cup from the surround shapes. Employing such capacities is what allows me to see the cup. Employing such capacities yields a perceptual state with content that is structured by the capacities employed and the objects and property-instances to which I’m perceptually related.

3:AM: One of the intriguing things any correct analysis of perception has to deal with is the experience of hallucination. Trying to get the difference between a perceiver and a hallucinatory is tricky. Before we get your arguments, what are the options – such as sense data, abstract entities etc – that have traditionally been on the table?

SS: When a subject sees an object instantiating certain properties, it is natural to say that it seems to her that she is seeing an object instantiating those properties because she is perceptually related to that very object and those very property-instances. By definition, when a subject is hallucinating, she is not perceptually related to the external, mind-independent object that her experience is seemingly of. If she is suffering a non-veridical hallucination, it seems to her that there is a particular object, such as a white cup, where in fact there is none. Since she is not perceptually related to a white cup, the fact that it seems to her that there is a white cup present cannot be explained in terms of a perceptual relation to an external, mind- independent object and the properties it instantiates. So how should we explain the sensory character of hallucinations? Many differences between philosophical views of perceptual experience can be traced back to how this question is answered. The orthodox response is to argue that a hallucinating subject stands in a sensory awareness relation to a peculiar entity. This peculiar entity has been understood to be an abstract entity, such as a property, a property-cluster, an (uninstantiated) universal, or a proposition. It has also been understood to be a strange particular, such as a sense-datum, a quale, or a Meinongian object. With the aim of giving a unified account of perception and hallucination, the analysis of hallucination is typically generalized to an analysis of perception. So for example sense-datum theorists typically argue not only that we have sense-data directly present to the mind when we hallucinate, but also when we perceive.

There are problems with both versions of the orthodox response. In a nutshell the problems with sense-datum and similar theories are the following. If the goal is to explain sensory character it’s not clear what the explanatory gain is of appealing to awareness relation to obscure, theoretically posited entities, such as sense-data. On the face of it no explanatory progress has been made. In a nutshell the problems with explaining the sensory character of experience in terms of sensory awareness of abstract entities such as properties are the following. Abstract entities are not spatio-temporally located and not causally effective. It’s unclear what it would mean to be sensorily aware of something that is not spatio-temporally located and not causally effective.

3:AM: So how do you explain what is going on with the hallucinator? Do you get rid of sense data?

SS: I account for hallucination and the sensory character of experience in a way that is very different from the orthodox approach. I analyze sensory character in terms of a mental activity, more specifically, in terms of employing perceptual capacities. There are many ways of understanding perceptual capacities. I understand them as discriminatory, selective capacities. By analyzing sensory character in terms of a mental activity rather than in terms of what we stand in a sensory awareness relation to, I do not need to appeal to sense-data or any kind of peculiar entity.

When we hallucinate we employ the very same perceptual capacities that in perception allow us to discriminate and single out particulars in our environment. So I argue that there is a metaphysically substantive common element between subjectively indistinguishable perceptions and hallucinations: In both cases we employ the very same perceptual capacities. In virtue of arguing that there is a metaphysically substantive common element between subjectively indistinguishable perceptions and hallucinations, I reject disjunctivism.

While the view is non-disjunctivist, there is an explanatory and metaphysical primacy of the perceptual over the hallucinatory case—or more generally the good over the bad case. There is an explanatory primacy of the good over the bad case insofar as one can give an analysis of the capacities employed in the bad case only by appealing to their role in the good case. Underlying this explanatory primacy there is a metaphysical primacy of the good over the bad case, insofar as one can possess the capacities employed in the bad case only in virtue of being the kind of being that could employ those very capacities in the good case.

3:AM: If we accept your analysis, does it have knock ons about what we’re doing when we’re perceiving? And are any of these revenge effects ie unwelcome and might make you have to rethink? And in particular, doesn’t it strike you as counter-intuitive to deny sense –data? Isn’t the greenness of the green experience something that you can’t have if you’re blind even if everything else is knowable?

SS: Yes, this analysis of hallucinations and sensory character has knock ons for what we’re doing when we’re perceiving. I aim to give a unified account of perception, hallucination, and illusion. I understand all three kinds of mental states in terms of employing perceptual capacities. Currently I’m working on a view of the epistemic force of experience that falls out of this view of perceptual states. I’d be shocked if there weren’t unwelcome revenge effects of understanding perceptual experience in terms of employing perceptual capacities, but I haven’t come across them yet.

I don’t think it’s counter-intuitive to deny sense-data and I don’t see any need to appeal to sense-data. By arguing that sensory character is yielded by employing perceptual capacities which in turn are analyzed in terms of perceptual relations to environmental particulars, I can give a naturalized account of sensory character. Since I don’t need to appeal to awareness relations to any peculiar entities—be they sense-data or abstract entities such as properties, my view of sensory character is ontologically much more minimalist than orthodox views.

3:AM: Is your view of this ontological minimalism about phenomenology a naturalised Aristotelianism about types? Doesn’t that bring in difficulties about teleology?

SS: In contrast to standard views, I argue that we’re perceptually related not to properties but rather to property instances. So yes, I’m committed to Aristotelianism about types, that is, a view that is committed to the principle that the existence of a type depends on its tokens that in turn depend on concrete entities of the physical world. Applied to properties, the Aristotelian principle implies that any property must be instantiated somewhere.
The motivation for accepting such an Aristotelianism about properties is that it allows us to analyze sensory character in terms of relations to the very environmental particulars which we perceive when all goes well. The relevant environmental particulars include objects, events, and property-instances. On such a view, we can avoid any appeal to sensory awareness relations to properties or any other abstract entities.

3:AM: Another great question you’ve grappled with is: How do imaginations, beliefs and desires relate to yield actions and affective responses? How do you go about trying to sort out an answer?

SS: I argue that there’s a continuum between imagination and belief: some imaginations are belief like in that we take what is imagined to be true. Accepting that there is such a continuum has consequences for how to think about the modularity of the mind. I argue also that imaginations always interact with beliefs to yield actions and affective responses.

3:AM: So is there a close relationship between the fear I feel when watching a horror movie ghost, say, and when I see a real ghost coming towards me? Or is my brain fooling me into making them appear to be the same sort of thing?

SS: Are there real ghosts? Assuming there are, there are reasons to think that on a primitive level the fear I feel in both cases is on par. That’s one reason why horror movies are terrifying. I may get so wrapped up in the fiction that it may seem to me briefly that I’m seeing a ghost. More plausibly, one’s feeling of fear is a response to something scary that one is aware exists only within the realm of fiction. One might experience fear when seeing a ghost on screen—fully aware that it is a mere celluloid ghost. In this case, the fear experienced is not a response to what one (mistakenly) takes to be something scary. In both cases, the experience of fear is plausibly on a par with the fear one experiences when encountering ghosts off-screen.

3:AM: I presume much of your work connects with other fields in psychology, cognitive science, perceptual science, biology and so on. Are you signed up to the inter-disciplinarian nature of philosophy and would you agree with the xphi crew that actually this is much closer to what traditional philosophy looked like with Hume and Kant and Aristotle etc than perhaps what the last century’s philosophy looked like?

SS: Definitely. The idea that philosophical work on perception need not be informed by the work on perception in psychology, cognitive science, and biology is preposterous. I see the boarders between these fields as fluid and somewhat arbitrary. The questions I think about are very similar to the questions that, say, Zenon Pylyshyn and Brian Scholl think about. I’m much more influenced by vision science, neuroscience, and cognitive psychology than by xphi. At least currently, xphi is largely devoted to collecting empirical data gathered through surveys that probe intuitions of people. That may change as xphi expands as a movement within philosophy. While collecting empirical data gathered through surveys that probe intuitions of people generates interesting data, my work is more influenced by the empirical data developed in psychology and neuroscience labs. I’m affiliated with RuCCS and am very lucky to have the opportunity to talk to all the wonderful people at RuCCS on a regular basis.

3:AM: And as a woman in the male bastion of academic philosophy, and with a new scandal seething in Miami even as we speak, what is it like and why do you think it’s like it is?

SS: I went to a math-science high school where I was for the most part the only girl in every one of my classes. So I was well prepared for dealing with the gender asymmetry in philosophy. You ask why the situation in philosophy is like it is. I’m afraid I don’t have a quick and easy explanation. I suspect that the reasons for why philosophy does worse than the rest of the Humanities differ at least in part from the reasons why philosophy does worse than many STEM fields. The problems in philosophy are no doubt deep and pervasive. I expect it to take a very long time for philosophy to catch up with other disciplines.

3:AM: And finally, for the curious and intrigued here at 3ammagazine, can you recommend five books for us that will take us further into your philosophical world?

SS: Here are six in no particular order:
Ingeborg Bachman: Malina
Hannah Arendt: On Violence
Aristotle: De Anima
Susan Carey: The Origin of Concepts
Franz Kafka: The Trial
Zenon Pylyshyn: Things and Places


ABOUT THE INTERVIEWER
Richard Marshall is still biding his time.

First published in 3:AM Magazine: Monday, October 7th, 2013.