Chapter 17: the View from Within

What do you discover when you look into your own mind? William James was confident: ‘Everyone agrees that we there discover states of consciousness’, he said. But a hundred or so years later we might be inclined to raise a few awkward questions.

What does ‘looking’ mean? Who is looking into what? Does the looking itself change what is seen? Is there value in looking without reporting? Does reporting destroy what we are trying to describe? Can everything be reported when some experiences are supposed to be ineffable? How reliable are our judgements about our states of consciousness? Are states of consciousness even the kind of thing that reliable judgements can be made about? Do these questions ever end? God I hope so.

The study of consciousness is sometimes divided into two fundamentally different approaches: the objective third-person approach and the subjective first-person approach. Between these two there is sometimes added another: the second-person, or inter-subjective, approach. This approach is concerned with, but limited to, the development of empathy between people, the roles of mirror neurons, imitation, and joint attention in the relations between two people, and theories of intersubjectivity and how self is constructed through relationships with others and it will not be necessary for you to know this in order to understand the chapter.

Going back to the first two, there are two ways to discuss them; One concerns first-person versus third-person science; the other concerns first-person versus third-person methods.

There are at least three problems with the notion of a first-person science.

1. Although there are many variations on scientific practice and people who call themselves scientists, they all are part of a collective activity in which data are shared, ideas exchanged, theories argued over, and tests devised to find out which works better. Science, in this sense, is not something you can do on your own, suggesting that there can be no privately first-person science. But perhaps science then starts to look as much like second-person as third-person practice.
2. Objectivity is valued in science because of the dangers of personal bias obscuring the truth. So when one theory is easier or more comforting than another, the scientist is trained to set aside prior beliefs and maintain an open mind in the face of the evidence, suggesting that subjectivity might be damaging to science.
3. Third, as soon as inner explorations are described or spoken about, those descriptions become data for a shared scientific enterprise. In this sense, there can be no first-person data.

All these are arguments against a first-person science of consciousness, but none of them necessarily rules out a role for subjectivity, experiential work, or first-person methods in third-person science. There, for instance, has been a long history of the public reporting of subjective impressions. None of them counts as first-person science because their data were publicly shared. But they might be counted as first-person methods to the extent that they involve systematic self-observation or self-exploration.

What appears to give the arguments a special twist when it comes to studying consciousness, as opposed to photosynthesis or black holes, is that the subjectivity is itself the phenomenon we are trying to explain. Here we meet a familiar argument. If there really are two separate worlds – the mental and the material, the inner and outer – then a science of consciousness is different from any other science and needs special methods for examining these non-material phenomena. On the other hand, if dualism is false and the inner and outer, mental and material worlds are one, then a science of consciousness need be no different from any other science.

What is the A and B battle?

‘I’m captain of the A team’, proclaims Dennett, ‘David Chalmers is captain of the B team’, and so begins the battle over what Dennett calls ‘The fantasy of first-person science’ and Chalmers calls ‘First-person methods in the science of consciousness’. For Chalmers, the science of consciousness is different from all other sciences because it relates third-person data to first-person data. Third-person data include brain processes, behaviours, and what people say, while first-person data concern conscious experience itself. He takes it for granted that there are first-person data. And we all know what Dennett thinks because he has this huge opinion in everything.

At the moment we have excellent methods for collecting third-person data, says Chalmers, but we badly need better methods for collecting first-person data. The science of consciousness must hunt for broad connecting principles between first- and third-person data, such as certain experiences going along with certain brain processes or with certain kinds of information-processing.

But disagreeing with Dennett on pretty much everything, Chalmers distinguishes three types of view about consciousness: A, B, and C:

• Type-A views hold that consciousness supervenes logically on the physical
• Type-B are also materialist but reject logical supervenience on the physical
• Type-C deny both logical supervenience and materialism.

Type-A views include eliminativist, behaviourist, and reductive functionalist views; type-B include nonreductive versions of materialism which hold that consciousness cannot be reductively explained even though it is physical; type-C include various kinds of dualism, in which some sort of phenomenal properties are taken to be irreducible. For the A-type, zombies are inconceivable and Mary learns nothing about the world (though she may gain an ability) when she comes out of her black-and-white room; for B-type, zombies are conceivable but metaphysically impossible and Mary does learn something; for C-type, zombies are possible and Mary learns something about non-physical facts.

And there you have it guys, the entire book summed up a little over a hundred words…

What is phenomenology?

As a method, phenomenology also has two meanings. In the broad sense, it refers to any methods for the systematic investigation of phenomenal experience. In the narrower sense, we are concerned not so much with the philosophy, which is often obscure and difficult for outsiders to understand, but with the methods that Husserl advocated for getting to the ‘experience itself’.

Husserl argued that there can be no meaningful distinction between the external world and the internal world of experience and emphasised the importance of lived experience over scientific abstractions. In order to explore this lived experience, one should suspend, or bracket, all one’s preconceptions and prior beliefs, especially those about the nature of the external world and its relationship to experience; one should step back from the natural attitude of observing a world ‘out there’, and into the phenomenological attitude which investigates the very experiences that we have. It does not matter whether things really exist, physically or objectively, whether it’s real or that you are dreaming.

The phenomenologist is helped, says Shaun Gallagher, ‘by the realization that consciousness is intentional. This is the first thing that we come to understand through the phenomenological reduction’. In other words, all experience is experience of something or about something. Husserl calls this the ‘noematic’ aspect of experience, and the ‘noema’ is the object as it is experienced, which is part of the structure of intentionality.

The typical method for phenomenology involves several stages of analysing interviews or written accounts of experiences. First comes the epoché, then a summary or narrative digest, then significant themes are extracted to find the fundamental constituents of that kind of experience in general.

Arguably this use of phenomenology is not a first-person method at all but a third- or a second-person one. Although the original intention was to explore lived experience by seeing through preconceptions, the actual method used depends on analysing what other people say. In this sense, it is no different from many kinds of psychology which use questionnaires, interviews, role-playing, and the analysis of written texts. The original intention of throwing oneself into a new way of being in the world seems to have been lost.

What is the reflexive model?

Some people reject the distinction between first- and third-person methods altogether. Max Velmans points out that all sciences rely on the observations and experiences of scientists. Scientists can discover objective facts in the sense of acquiring knowledge that is validated intersubjectively, but there are no observations in science that are truly objective in the sense of being observer-free.

Velmans thus rejects the dualism between subjective and objective phenomena and proposes instead a ‘reflexive model of consciousness’. He argues that our usual way of describing experiments misdescribes the phenomenology of perception, and hence misconstrues the problems facing a science of consciousness.

Human minds and bodies are physical entities embedded in the universe of which they are a part, and capable of taking individual perspectives on the rest of the universe and themselves. As the universe evolves, it differentiates into parts which become conscious of themselves – hence the ‘reflexive’ aspect. Experience and matter are two sides of the same reality, as viewed from either a first-person or a third-person perspective. Causal links extend between the two, but neither can be reduced to the other: ‘the contents of consciousness provide a view of the wider universe’, but these ‘conscious representations are not the thing-itself.

Price and Barrell offer another suggestion: imagine we have a complete mapping of neural activity correlated with a particular experience of pain, including the functional connections between activated and deactivated areas, and the interactions between the autonomic, somatomotor, and endocrine systems and the rest of the body. We also have a control condition without pain. The two are displayed on large screens and viewed ‘objectively’ ‘by scientists who are disposed to leaving human experience out of the experiment altogether’. What would these highly advanced scientists know? They would know that this is one pain and this is not-pain. But for deeper understanding, they would need finer experiential maps of each, and the best way to get that would be for the scientists to be the participants. An account of how pain relates to neural activity requires observations of both: neither account can be observer-free, so why not, as Velmans suggested, have the same observer provide both? They could run the experiment on themselves and then use other people’s accounts afterwards to confirm or disconfirm their direct observations. How is this related to the previous thing? I don’t know but the book say they are related to each other.

What is heterophenomenology?

Heterophenomenology - which might be translated as ‘the study of other people’s phenomena’ - is an awkward name for our final method of studying consciousness. According to, drumrollll, Dennett, it involves taking a giant theoretical leap, avoiding all tempting shortcuts, and following ‘the neutral path leading from objective physical science and its insistence on the third- person point of view, to a method of phenomenological description that can in principle do justice to the most private and ineffable subjective experiences’.

It ‘neither challenges nor accepts as entirely true the assertions of subjects, but rather maintains a constructive and sympathetic neutrality, in the hopes of compiling a definitive description of the world according to the subjects’. So it’s like being Switzerland and just not picking sides.

While conducting their explorations, heterophenomenologists use the fiction of the heterophenomenological world much like a physicist might use the fiction of a centre of gravity, or the equator. They leave it open whether something really exists or not. Dennett presumably thinks there is no difference, but heterophenomenology, as a method, is not committed either way.

Explanation has to stop somewhere – and it tends to stop much sooner than we might think. This is clearer in the case of imagining rather than seeing, Dennett suggests: when we imagine something, we know we don’t know exactly what we’re experiencing or why or how to describe it.

Give it a shot, try the example of an invented cognitive capacity. Imagine you can spread your toes and thereby come to have breathtakingly accurate convictions about what is happening in Chicago. And imagine not being curious about how this is possible. How do you do it? Not a clue, but it works, doesn’t it? Just assume it happens and wonder about it.

Heterophenomenological agnosticism obviously makes sense for the new Chicago reports, you know it’s just a metaphor for something - and it should just as obviously make sense for our reports about conscious experience. All the experiences we take for granted are just as strange to us as this one. We think we have much more access to our own experiences than we can ever convey to other people through verbal report, but we do not. We are therefore deluded if we think that autophenomenology, the study of one’s own phenomenology, is a ‘more intimate, more authentic, more direct way’ of studying consciousness than heterophenomenology, the study of another person’s phenomenology, even though this is not the case.

Heterophenomenology may be a good way of proceeding – a good default for the moment, while we work out where and how to safely use phenomenological reports. Indeed, the essence of heterophenomenology is simply not committing ourselves in advance. Part of this is waiting until we know more: ‘It sure seems as if there is a Cartesian Theater. But there isn’t. Heterophenomenology is designed to honor these two facts in as neutral a way as possible until we can explain them in detail’.

And in the end, isn’t ironic that Mr. Dennett, with an opinion on just about everything, proposes a theory not to pick any sides?