Antony is interested in naturalistic epistemology, and is currently attempting to create a psychologically realistic accounts of empirical justification. She works in biblical doctrine, coediting with Charlotte Witt A Mind of One’s Own: Feminist Essays on Reason and Objectivity.Antony continues to be a guest on the radio show Philosophy Chat, and has contributed to The New York Times. She was recently elected Vice-President of the Eastern Division of the American Philosophical Association, and has served as President of the Society for Philosophy and Psychology.
Zan Boag: I’d love to delve into your own work as a philosopher of mind. Obviously, that is not the only thing you’ve done, but I’d like to find out about the overlap between philosophy of mind and cognitive engineering. Can you operate with cognitive scientists?
Louise Antony: I don’t do some original research in cognitive science . I came of age as a philosopher in the’70s. I was at Harvard from’75 to’80. And that has been right around the time that the cognitive revolution was happening and has been directed by people at MIT, especially Noam Chomsky and, on the philosophy side, Jerry Fodor. I hung out at MIT and took courses and seminars there as far as at Harvard. And in reality, Harvard as a department at that time, was very antagonistic toward cognitive engineering. It was a sort of Neo-Wittgensteinian department in a lot of ways. Hilary Putnam was my official advisor, but afterwards having articulated a functionalist conception of the head and normalising the idea that higher-order sciences like psychology were not likely to be readily reducible to biological kinds — a mental kind, Putnam argued, would be multiply realisable in different kinds of substance. Now, the only minds we know about are the heads which are embodied in brains, but there might be heads embodied in other sorts of bodily systems. Jerry Fodor took this idea and ran with it. However, Putnam bailed from the concept that philosophy of mind should be connected to cognitive science early on. I had been in the middle of an antagonistic place, which I believe was, for me personally, very stimulating and fruitful.
I actually was in the thick of the cognitive revolution. The foundation of the Society for Philosophy and Psychology happened during the time that I was in grad school. I eventually became the president of the society. So here is the sort of irony: I had been a student of Willard Van Orman Quine, who said that we needed to naturalise epistemology and the thought was to start asking how we figure out how to get to the theories that we get to from the data that we have offered to us. That seemed to me to be an invitation to do cognitive science; not to Quine!
The Society for Philosophy and Psychology was great because it is an interdisciplinary society which tries to attract psychologists and linguists — in the early days there have been more computer scientists than there were recently, and ecologists — as it became OK to talk about beings with thoughts, people started to say,”Well, animals appear to have minds also.” With the passing of the behaviourist paradigm, the analysis of animal minds took off and there has been fascinating research coming out concerning the arrangement of animal minds.
A lot of people are a bit uncomfortable talking about animal minds because in doing this, we have to acknowledge that we’re animals ourselves. I believe humans have a hard time admitting that we may be just like any other creature. We seem to have positioned ourselves at the peak of the tree, so this idea that there are different minds is unfathomable to a people.
You’re talking the notion of other thoughts in the’70s however, were you?
Yes, undoubtedly. The primate research is extremely compelling. It’s complex and it needs to be done carefully, but psychologists like Laurie Santos at Yale, and some of the folks doing work in the Yerkes Primate Center. Folks have pets and they attribute intentional states for their pets that I don’t think they’re wrong to do this. We had a puppy, Freya. Freya was not the smartest dog in the world, but there was this one time she clearly needed a plan. She’d been removed to urine by my husband first in the afternoon and she found a doughnut that someone had likely thrown out a car window or something. And she lunged for it and Joe grabbed her and prevented her from eating it.
I didn’t know this had happened, but she spent the rest of the morning calculating how to get herself back out to this doughnut. You can practically see the gears turning like,”OK. I must find outside. How can we get outside? I must find the leash. How can we get the leash? I must get the human. Which human is it? Okay, it’s Louise. All right. I must get her to take…” You understand what I mean? And it is rather difficult to explain their behavior if you don’t create those mentalistic attributions. The same as your dog, they are ruled by their gut. I think they’re coming up with a plan to attack the new hedge I’ve planted, but actually they’re drifting ago, and they see it there, and also their stomach tells them”Hey, I need to eat this.” Perhaps it is the same with the dog also. It was actually only hunger pushing her back .
I do not think the important question is what is the ultimate goal because I get hungry around 5:00pm, right? That’s a biological force. The question is what capacities can I recruit to accomplish that objective. People and lots of different animals have this cognitive capability. We can represent our objectives, and we can represent possible means to the aims, and we could decide between them. So it’s not really a question of what the ultimate aim is. I mean, with having sex: how am I going to organise things so that I get some tonight? The goal is biologically given, however, the means are subject to all sorts of calculation.
It’s not a question of where the target comes from. It is a question of exactly what are your options for realising the goal? And I believe if you’ve something like, mosquitoes, ” I do not think they compute. I really don’t believe they say,”Oh, there is a dinner party on the market. There are most likely to be some meals on the market.” There is some physical invariant, I think that it’s the smell of perspiration or something like that, which activates a dedicated pathway to movement that allows them to match their goal. I think the distinction is if there’s something cognitive connection between the goal and the behaviour that’s taken to realise it.
So the mosquito has been reactive and the dog has been proactive in its own activities? Yes. The dog is cognitive. However, for most of us, nearly all of our lives, the biological demands that we have — we encounter some cognitive mediation regarding how we are going to begin realising them. Occasionally we choose not to eat. That doesn’t happen to me very often, but people go on hunger strikes, or individuals diet. Even breathing: there are plenty of fundamental biological processes which we are able to take voluntary control of, or to an extent. Hence the chance of cognitive mediation, I believe that’s likely a natural fracture in nature and I think it probably goes fairly far down the phylogenetic series, if you would like to think of matters in a hierarchical way.
When we had been connected before, you’re speaking about the distinction between the scientif ic and everyday interpretations of these terms’perception’ and’perceive’. Each interpretation gives you a different launching pad from which to answer questions. So. If you are considering it from a scientific perspective, what is perception?
I think the scientific issue is,”How does our interaction with the environment generate information that may be integrated with our other cognitive conditions?” In some organisms, perception is linked very, very tightly to behavior. And there are some aspects of human perception, reflexes we call themwhere that is the situation. Now, our perceptual systems like, enjoy vision… vision, I believe, is probably the most wellunderstood perceptual system, possibly audition then. There’s a very rich body of research on vision and the fundamental problem is that the stimulus vastly underdetermines the representation we finally construct. We know what the physical interaction is, we know that light irradiates our mind. The retina is a two-dimensional grid. And from patterns of activation and that two-dimensional grid, we generate a representation of this environment that’s extremely rich.
For one thing, it’s a representation in 3 dimensions. So we know a couple of things. We are aware that there has to be some process that enriches the informational input that we must empower us to inform that items are arrayed in three dimensions in front of us. I meanour visual phenomenology is that the world is in three dimensions. We’ve got a variety of different mechanisms for figuring out how things are arrayed in three dimensions. One of them is ocular disparity. And you will find computational mechanisms that can compute the disparity between these pictures and actually tell us what that means about where things are in three-dimensional area, but that only works, I have forgotten just how far off, but it’s only 20 to 40 ft )
However an awful lot of our depth perception is a matter of calculation on the basis of what some understanding theorists call”hidden assumptions”. So it looks like we’re primed to draw certain inferences from specific patterns of input. If we view, say, a line of trees like the visual impression… I mean, in the event that you just looked at the picture that is painted on the retina, the lines of trees seem like they’re converging at the horizon and the peaks of the trees are getting smaller and smaller as they go toward the middle of the visual area. We’re natively primed to respect that as an indicator of thickness.
And that is something which does not depend on our getting two eyes. A little anecdote: we had a cat for some time who turned out to have cancer in one eye and had to have the eye removed. I asked the vet, I said,”Geez, the cat had been jumping around doing things. Is not this going to be a huge detriment to her ability for a cat?” The feline-canine ophthalmologist said,”I believe ocular disparity is overrated. I really don’t think cats or dogs or anybody really relies on it. I think if there is an evolutionary adaptationist accounts of why while we have two eyes, it’s redundancy. So when you get caught in the eye with a rod, you still have a different eye.” And in actuality, the cat, over four to six hours of the surgery, was leaping up and down out of stools. I mean, only acting like everything was absolutely fine.
So anyway, the moral of the story is we have a good deal of mechanisms for depth understanding and we use them all. That is a way in which we can add info into the raw perceptual data, which is just in this case the irradiation, the patterns of light on our retina. What do we understand about the information coming in? We can normally figure out that, we know that in the case of all of the senses, I think. And what exactly do we get at the end? What sorts of computation, what kinds of additional info do we have to posit in order to explain how we get from the material that comes into the perceptual experience that we have at the end? We can not always trust what we see. One example is that the moon around the horizon appears much larger than when it’s overhead. There are lots of optical illusions, such as the MüllerLyer illusion. Although we take in this information, we can’t always anticipate what we see.
There is a debate in vision science about the level to which the perceptual processes involved with any particular detector modality are informationally isolated from different stuff. And particularly, informationally isolated or encapsulated from background information that we have. The view I subscribe to — I am not a vision scientist, but based on my reading of this literature and the philosophical problems that are linked to it, I’m convinced that sensory processes are automatic and manifold. Take eyesight , eyesight is going to develop a decision about the way things appear. And that gets passed to higher cognition and higher cognition may take it or leave it. Greater cognition can say,”OK, you’re telling me, it looks like there’s an elephant there, but we are in Massachusetts and that I do not really think there are elephants there. I think something else has to be going on.” But vision is stating,”I am just telling you exactly what it looks like. It looks to be an elephant.
And I think from the perspective of epistemology, the point of view of concerns about justified beliefs about the outside world, I believe that film makes an awful lot of sense. The job of every individual sense modality is to tell us how it sounds based on the proprietary information for that sensory modality. So the examples that you just mentioned, the Müller-Lyer illusion, visual illusions I think provide a lot of evidence for this total picture of the connection between perception and cognition, because eyesight is stating,”How it looks to me is that those lines aren’t equal. I am just telling you how it looks.” The intriguing thing, the really important thing, and this is something which was emphasised by Jerry Fodor and by vision scientists, is the illusion does not go away when cognition simplifies it.
Now you can find other things. I’ve got a peculiar view at the epistemology of perception. I think that the job of perception is essentially to bring the outside world into the mind. I think what understanding does is create a trusted record of the external world. It can just give a trusted record of the external world, as it were, from its vantage point. So eyesight can only let you know what things look like, and it is likely to agglomerate disparate, distal situation that look exactly the same. So if my dog is lying on the grass and generates a specific visual pattern, then that visual pattern may be visually indistinguishable from a weird pattern of shadows on the grass. And vision is going to say,”It’s this Tiggyish pattern” I don’t understand if Tiggy’s there or not. I can not tell that. I can tell is how things look. And also the Tiggyish pattern may be Tiggy or might be this pattern of shadows. And that’s not perception’s job. It’s cognition’s job to sort things out.
When it comes to perception, vision is not working alone though, can it be?
So it’s part of my religion that it is. Vision’s working alone; audition’s working alone. Cognition has to figure it all out.And it’s because we’ve got this cognitive interface. If I were standing close to you, physically, and I were to go like this with my finger, you could believe with all your heart that I’m not going to poke your eye out, but your eyes are going to blink. That’s a hardwired connection linking between a certain pattern of visual perception and a certain behaviour. That’s a reflex. But most of the way perception works in our cognitive life is not reflexive. It goes through cognition. It goes through some consideration of what’s actually true.If vision is working alone when it comes to perception, what happens to those without vision; what happens to the way they perceive things?Well, they have other modalities. I don’t know a great deal about audition, but audition also reflects objective structures in the world. The frequencies of light that affect our retina are relatively short compared to the frequencies of sound that affect our ears. And that means in a certain way that vision can make finer distinctions than audition can make.An organism has to deal with whatever information they’re capable of getting from the world. I just don’t know what the limits are… I think it’s a completely empirical question: what amount of information can be gleaned from the world with just one or another sensory modality? I mean, there’s this very interesting research on echolocation among the blind. Blind people do use tapping of their canes to get information about the physical layout of the environment. There was one case, it was profiled on the extremely scholarly forum called This American Life, the radio show. The subject of the story was incredibly proficient at using auditory information to figure out the layout of his environment. He had had some vision as a child and as a young child, but he had to have both of his eyes removed because of disease. I’m not sure how old he was. I don’t know if that early visual experience played some role in his being able to echolocate. One question is to what extent is visual cortex – which is the part of the brain that analyses visual input – dependent on patterned experiences early on? There have been horrible experiments done with kittens that cut the connections between their visual cortex and their eyes, and they turn out to have all sorts of deficits about being able to orient themselves physically, even by touch in their environment.It looks like the computational centres in the brain that are responsible for dealing with sensory input may need, as a causal factor, certain kinds of input to develop the computational mechanisms to deal with the influence. So this particular individual was not blind from birth, he did have some visual input and that might make a difference, but he developed an ability to echolocate to a very impressive extent. I believe he has also trained people who are adults who are blind to use echolocation. Of course, there’s information you can’t get. You can’t get information about surfaces from bouncing sound waves. You can’t get information about colours. But it looks like there might be cortical structures that are primed to receive input that can form certain patterns.And there’s also this interesting research on sensory substitution. This is where you take something like the tongue that has lots of sensors and you put a patch on the tongue that’s connected to some kind of photo sensor so that patterns of radiation are conveyed to the tongue. And, over time, blind people can learn to construct models of their physical environment that are accurate and very useful. And they even report some visual phenomenology about that. I haven’t really delved into that. I mean, there are other philosophers who have looked at that. Jonathan Cohen is a philosopher who has looked at that.That’s interesting because I wanted to ask you about Maurice Merleau-Ponty’s theory of embodied perception. Our bodies interpret the world in different ways – it is through our bodies that we make sense of the world around us. How does ageing affect how we perceive, or interpret the world around us?I think this sort of commonsense use of the term perceive makes sense of it in light of what we experience. There are an awful lot of factors that go into what we perceive in this more informal sense. Two important considerations are focus and interest. So I’m 67. I’m experiencing some orthopaedic problems. I do wear glasses. I have graduated lenses. I love to hike, but I find that I have to turn my head straight down so that I have the right part of my lens focused on the part of the path that will tell me whether I’m going to trip over a root or a rock or something like that.So there are certainly accomodations we have to make to our changed sensory relationship to the world and there are also accomodations we have to make based on the behaviour of our muscles. I’m an amateur musician. I play the flute and the piccolo. And, so far, despite some tendonitis in my finger and some arthritis in thumb, I have not found that my playing has been particularly impaired. But I do like to hike and my mobility has been impaired a lot. Now, I pay attention to things that I didn’t pay attention to before. I mean, I would just sort of plough through the woods before and now I’m very, very careful about where I place my feet, and I found myself planning about how I’m going to get down this particular steep part. So we can direct perception and we can direct it in light of our goals. That’s one factor that we have control over.
A variable that we don’t have control over is if the perceptual, muscle pathways that were established when we were younger are still undamaged. Speaking back again, there were movements that were perfectly fluid for me personally I would not think twice about doing this I now find myself sort of inhibited about, and I have to kind of strategy. There is a very complicated interplay between the established perceptual muscular pathways and exactly what occurs if those begin to degrade, or deteriorate, or alter. But going back to songs, I believe there are pathways which begin as mediated. When I was first learning how to play the flute, I would look at a note published in musical notation, I would say,”OK, that’s a D so that is this finger, that finger, this finger, or this finger, this finger.
And I meanit would be a rather conscious thought about the way I think of the note. There is no conscious cognitive mediation any more for me about the majority of the notes. I just see what is on a webpage and finger them correctly. And it is apparent that there’s cognitive mediation at the same point that subsequently becomes automatic so you do not have to consider it any more. And I think that that’s probably involved in all of the skilled bodily behaviours that we acquire. There are most likely lots and lots of patterns that are kind of specified that we just… walking for many human beings…
I wonder if this automation goes to how culture and gender has shaped us. The way we act, the way we hold ourselves, the way we interact with others is formed by the particular culture which we’re born into, for example, gender roles that every culture can assign to us. We also do have for certain phases of our lives, distinct bodies. I think in the start, in the long run, they are probably quite similar, but there is a stage in the centre where they’re different and they perform different functions. At particular times, especially when you’re taking a look at something such as breastfeeding and childbirth, these sort of things, just 1 body can do that. So how do culture and gender shape the way we perceive the world? And again, I’m speaking out of a metaphorical perspective, not in a scientific perspective.
Well, I believe there’s massive cognitive mediation. I believe one of the profound ways that sex roles shape bodily experience is through attention, what you listen to, what you look out for, everything you’re concerned about. And again, when I’m referring to gender here, I have in mind female-sexed bodies for the most part. I don’t presume to understand what the experience is of trans individuals, but I do understand growing up as a cis heterosexual lady, that awareness of menstruation, awareness of this liability to pregnancy was something which was in my mind constantly during adolescence and early adulthood, and it continued to function as I grew older, but I just had more control over my choices.
I don’t know phenomenologically what it’s like to develop as a cis male. I don’t think this sort of shaping is essential, but I do not think it’s philosophically very significant since it is all part of the overall truth that the perspectives you come out with are likely to be a use of the data that goes in. And unique bodies will get different data going in and that is likely to interact with their cognitive states in various ways. I mean, teenage heterosexual men will be worried about pregnancy in a different manner than adolescent heterosexual women will worry about pregnancy. I really don’t think it’s a major deal epistemologically because I believe that it’s an issue of looking at the information that’s coming in and also the interests and beliefs that you have.
You mentioned that you can’t understand what it is like to be a cis male. Thomas Nagel famously pondered what it was like to be a bat — even if we were to become bats with echolocation, insects, and all — which we would only experience the behaviours of a bat; we would not really know what it’s like to become a bat. I just wonder if this may be extended to all other minds, whether we’re ever able to understand what it’s like to walk into someone else’s shoes?
I’ve always had difficulty with Nagel’s article because I am not very sure what the finish of it’s assumed to be. Obviously, I can not understand what it’s like to be a violin, even if I become really proficient at echolocation, because my echolocation will produce mental conditions that interact with other psychological conditions that the bat does not have, or which are different from the bat, also it might be my echolocation is clunky, I’m not as attuned to slight differences in different frequencies of sound that I’m receiving as a bat is. So I think what Nagel’s post does is increase a general question concerning how we can produce any meaningful comparison of subjective experiences. I have profound disagreements with my husband about this. I think probably we do not know what it is like. We won’t know what it is like. I really don’t believe the blind people that will take part in echolocation know what it’s like to be a bat, since I don’t think the echolocation they participate in is incorporated into their behavioral and cognitive life in the manner that the bat is. But I believe if you say,”Well, then we couldn’t really know what it is like to be a bat.” I think it does take you in a sceptical direction where you have to say,”Can I know what it is like to have any other experience in common with any other human being?”
I’d like to return to’influence’ now. Historically, there has always been someone telling others how things ought to be translated, how life should be, what’s important, what’s right and wrong — telling them how to live and that to be. But now we have a media that is so pervasive that it influences us every hour of every day. Which websites we select will decide what we believe to be wrong or right. The question would be, to what extent does the media that we happen to consume determine the way we perceive the world?
One important independent variable is that media we pay attention to. A number of the men and women who encourage Donald Trump listen to Fox News and another Quinean Maxim is that you can hold on to some belief if you’re eager to make suitable adjustments elsewhere in your system. It’s not that hard to produce a coherent story about matters if you are ready to dismiss facts, but saying that, I mean, who do I trust about the data about COVID infections and deaths? I’ve got my sources. I trust The New York TimesI trust The Washington Post. I mean, there are other media outlets which I will visit and I will attempt to place them together and see if they all agree and so forth.
But the fundamental question about why do I trust the sources that I trust, I believe, is a really hard question and it’s very tough to reply in a formalistic way. I believe the paradox of the net was that it was going to be this huge democratising force. I worry about this much as a professor. I thought for quite a while about there needing to become a path in practical epistemology. There is this epistemology going on and we are supposed to be issuing norms for belief formation and things like that. Well, what is the advice we have to offer ordinary men and women?
People are kind of left to their own devices and the people who are working the longest hours, performing the heaviest physical labour, people who have the fewest choices about outsourcing and home cleaning and food preparation, etc, they are the most pressed individuals. They don’t have a great deal of time or leisure to be able to sift through a lot of competing narratives about the news. I believe late capitalism has created conditions where there is a kind of force against individuals being connected to reliable sources of information and sources of information that take their interests into account. I think that is a massive issue. Can we alter our understanding of the world?
Yes, I think we could learn things about our epistemic vulnerabilities. There’s a strong literature now on what’s called”implicit bias” on methods by which acquired patterns of institution can influence our behavior even if it doesn’t influence our considered judgements. I think that what we know about the formation of belief and the best formation of belief kind of gives a lie into a certain normative perspective about how we need to form belief, which I call Dragnet epistemology, following the series Dragnet. I believe the ideal that is embodied in that display is a radical empiricist opinion that says you just have your sensory input. You just calculate, on the grounds of this, to generalisations, and you test themand if they come off, then you think it. If they don’t, you then alter your belief. I mean, that’s a totally unrealistic image of how human cognition works. That is not the way things happen. I believe that dealing with the fact of the failure of the very appealing, but completely false image, of how we form beliefs is a part of developing epistemologically. And when you do grow up, it gets much, much more complicated. I still have the liberal hope that when we left education available to everyone, if we gave people enough stuff wellbeing, that they didn’t need to be concerned about where their next meal was coming from or where the roof was going to be over them, that they would have enough cognitive and emotional leisure to consider matters more carefully.