:: Article

All About the Ego Tunnel


3:AM: And then how do we get to present ourselves as being the centre of this reality?

TM: I have said many years ago that the self-model differs from all other models in that it is functionally anchored in the body though a persistent causal link. There is a continuous flow of information from the upper brain stem and hypothalamus, which presents, for example, the current state of affairs in terms of homeostatic stability – as Antonio Damasio once put it, emotions present “the logic of survival” to us. It is important to understand that there is not only “embodiment” in the broad sense of the term, but that there also is an interoceptive self-model: There are gut feelings, internal receptor systems in blood vessels, in tendons, muscles and joints, there is constant gravitational input through the vestibular organ creating a frame of reference. Interoceptive self-consciousness provides us with a much higher degree of invariance than the environment. The body model is grounded on these internal sources of information from which the brain can never run away, and which generally possess a high degree of reliability and which, as long as all goes well, provide us with a phenomenology of self-location in space, of permanence and stability over time. My speculative hypothesis is that what I call the “phenomenology of substantiality” is actually created by this low-level awareness of the very life process itself, the ongoing representation of successful self-sustainment. This then leads to theoretical intuitions that make it tempting to describe “the” self as ontologically self-subsistent, as a substance in the metaphysical sense.

3:AM: How does culture change the ego tunnel?

TM: An important question. Consciousness clearly is a culturally embedded phenomenon. But there are many readings of this claim. One concept about which I do not know enough, but which may be of great future relevance is that of a “cognitive niche”: Human beings construct cognitive niches, they are born into them, and they adapt to them. One beautiful new way of looking at the philosophical history of ideas, perhaps, is that can be seen as constructing the more abstract levels of humanity’s cognitive niche. Such self-constructed environments contain not only physical artefacts, not only representational systems that embody knowledge like writing systems, number systems or the specific skills and methods for training and teaching young human beings – I think they also contain conceptual tools with which we can ascribe properties to ourselves, eventually even changing the phenomenological profile of our subsymbolic self-model.

For example, it makes a difference if the concept of “consciousness” is available in your culture at all, if it is a pre-existing tool for cognitive self-reference, a semantic instrument that is offered to you by your linguistic community as it were, a potential scaffolding for social practices in which we can then all look at each other as “conscious” or “unconscious” beings. In 1988 the late Kathleen Wilkes published a paper in which she showed that in the very large majority of languages on this planet we do not even find an adequate counterpart for the English term “consciousness” (which only acquired it present meaning after what she called the “Cartesian catastrophe”). Why did all these linguistic communities obviously not see the need for developing a unitary concept of their own? Such question are one example of what “cultural embedding” can mean. If a culture develops the notion of a “person”, say, as of a “rational individual” and a “moral agent sensitive to ethical issues” – how does the availability and widespread use of such a term change the phenomenal self-model? How does it change its internal microfunctional profile, and what does it do the overt behavioural profile of its members?

Then there are other readings of “cultural embedding”. I have written a bit about how a subset of neurotechnology turns into phenotechnology, a “consciousness technology” that directly and primarily aims at changing the user’s phenomenology – think of virtual reality, robotic re-embodiment, or even molecular-level technologies like new psychoactive drugs. New scientific knowledge creates new technologies and new potentials for action. Market pressure and newly emerging cultural practices then begin to change the contents of consciousness itself.

That was one reason I founded a Neuroethics Research Group in Mainz a number of years ago: Neuroscience definitely needs a critical eye and a professional ethical assessment by philosophers. In Mainz we have mostly focussed on “cognitive enhancement”, new pills that purportedly make you smarter and more alert. I will not go into details here, but just add that after a decade or more in neuroethics my general conclusion is that many more proper analytical ethicists should move from philosophy into the new discipline of neuroethics. The issues are highly relevant (just think of pharmaceutical moral enhancement or new military applications), but from a philosopher’s perspective the level of debates is often a bit shallow and certainly has some room for improvement. It needs people who are not bound to some ideology or other and know proper analytical ethics well. This is a job for philosophers, and if we do not support policy makers and society as a whole by analysing and communicating the relevant options, then we may well get run over by new technological developments following on the heels of neuroscience.

How rapidly such developments can unfold has now become evident in the sphere of illegal psychoactive substances: In the English edition of “The Ego Tunnel” (which appeared in 2009) I cautiously predicted that the number of accessible illegal drugs could soon soar dramatically. In the expanded 2014 version of the book (not available in English) I already reported that in the three years following my prognosis, first 41, then 49, and in 2012 even 73 completely new synthetic drugs were seized in Europe alone; substances that had been completely unknown before. Now we see that the general trend underlying my prediction is unbroken: in the following year 81 novel psychoactive substances were seized for the first time, in 2014 the number was 101. If one looks at the respective annual reports of Europol and the European Monitoring Centre for Drugs and Drug Addiction, one is certainly justified in saying say that the situation is completely out of control by now. However, this observation now also applies to “cognitive enhancement”, off-label “brain-doping” with prescription drugs: As soon as a truly effective substance for boosting brainpower actually exists, even the most rigorous and strict forms of control aimed at their legal application will not work anymore. By now there are hundreds of illegal drug labs that would immediately copy the respective molecule and push it onto the illegal market. This is what you get for decades of denial, repression and systematic desinformation.

3:AM: Because your approach is representationalist and functionalist wouldn’t it be possible for something like the population of China, for example, or Google, to be conscious? If all conscious systems have to have phenomenal selves would we be committed to saying China had one, and Google?

TM: Phenomenal properties supervene on functional properties. If China or Google realized the relevant microfunctional properties that, for example, are instantiated by the minimally sufficient, global correlate of consciousness in our brains, then they would. We do not yet know what these properties are, and there is a possibility of principled mathematical intractability. But it is very clear that China or Google could never realize the time constants in neural algorithms, nor the complex, hierarchically nested dynamical profile of the conscious human mind or even the statistical physics making it possible. In 2016, asking such questions distracts from the really relevant issues. For example, the dramatic shifts in our notion of “representational content”.

3:AM: Dennett writes about the self as an illusion – but you deny this don’t you?

TM: First, the notion an “illusion” is an introduced technical term: As opposed to a misrepresentation on the sensory level in the absence of a stimulus – which would be a “hallucination” – it is a sensory distortion that actually has an external causal source, one that is internally modelled in an unusual way. As a matter of fact, many illusions may actually be “optimal percepts“from the perspective of an ideal Bayesian observer.

So “the self” could never be an illusion in that sense. Second, misrepresentations presuppose an epistemic subject, some entity that is wrong about something – but could in principle be right. But does this entity have to be “a” self? I am saying it is not a thing, but a process. The epistemic subject could be the person or biological organism as a whole, and it is well conceivable that such a system dynamically and fluidly operates under a conscious self-model without being phenomenally aware of this very fact. Many aspects of this self-model could be misrepresentational, shaped by evolution and society – unrealistic optimism, overconfidence bias, a robust misrepresentation of substantiality or personal identity across time, and so on. This also wouldn’t support trendy, omnipresent “illusion talk”. I must admit that I have occasionally been guilty of it myself, but it haunts me for years, just like the label “neurophilosopher” that journalists like to stick on me. Philosophy of mind is so much more than “neurophilosophy”!

As for the omnipresent “illusion talk”, I believe it keeps reappearing because people think: “Ah, I don’t really understand what all these present-day philosophers and neuroscientists are saying, but this stuff reminds me of something that I read in the New Age bookstore! Cool! Something that sounds politically correct and gives me a warm glow, something that sounds romantic, because it has something to do with all those popular books on Buddhism and Hinduism. Ego-death! Liberation! Giga-Bingo! Something pleasantly obscure and exciting that leaves a door open and makes all this scientific stuff a potential tool for death denial, something that confirms what I have always wanted to believe in!” Then people suddenly think you are a good person, and one of their own tribe.

But just begin talking about potential reductive explanations and the issue of intellectual honesty for a bit, then they suddenly say you are bad guy – a “neuro-nihilist”, boo! – what a negative and materialist version of reificationism, this is! You can see Evan Thompson’s new book for a recent version of this strategic innuendo. But I never believed the self was a thing. On page 1, the fourth sentence of Being No One already says “The phenomenal self is not a thing, but a process…” l must, however, say that really like Evan’s new book for many reasons, one of them being that he really endorses this basic idea and repeats the point over and over again.

3:AM: How do dreams help to show what is happening in the Ego Tunnel? Once we start thinking about dreams and wakefulness in your terms, doesn’t it make sense to ask whether we can have lucid wakefulness as well as lucid dreams?

TM: Dream research has always been a hobby of mine. My official position for many years has always been that the conscious wake state is a controlled hallucination, a form of “online dreaming”. Consequently, nocturnal dreams had to be offline states, and their relevance consisted exactly in the fact that, in humans, they seemed to be the only global states of conscious experience in which one could investigate what a complete subtraction of sensory input would lead to. In particular, I was always interested in what happens to the conscious body-model in such a functionally disembodied state, and I have looked quite a bit into the self-model in the dream state, unnoticed rationality deficits etc. But Jennifer Windt showed that that was wrong: Dreams are weakly embodied states, perceptual information actually does influence the dream state and the dream body in particular. I greatly recommend her Magnum Opus “Dreaming” to anybody interested in these issues – everything one needs to know in one single book, the best entry point into the debate in existence and probably for many years to come.


In 2007 we wrote a chapter on what actually happens to the self-model when, in a dream, we realize that we are dreaming. From the perspective of my own theory this is an important question: What exactly happens in the conscious self-model during the transition from an ordinary dream to a lucid dream? Is there a philosophically interesting form of “insight” taking place, or are we just dreaming that we are dreaming? Everybody has heard about the phenomenon of “false awakening” (you dream that you have just woken up, suddenly realize that you are still asleep), but could there be “false lucidity” as well? As a philosopher, I have written that it would be an important contribution on the empirical side of things if the neural correlate for dream lucidity could be isolated. German researchers Martin Dresler and Ursula Voss have actually done this, and much earlier than I thought it could even happen. Another issue that I have always found important is what I call “minimal phenomenal selfhood”, the question of what the simplest form of self-consciousness is. I think dream research is highly relevant here. Why? New considerations indicate that a spatially extended body image and also an interoceptive self-model are not necessary conditions for self-consciousness in the strong sense. An extensionless point in space as a unit of identification seems to be sufficient for a stable form of self-consciousness. This point comes out clearly in Windt’s immersive spatiotemporal hallucination model of dreaming and in a recent open access paper I wrote in which I have tried to give some answers to the question of Why are dreams interesting for philosophers? The debate now shifts to the question if perhaps even pure temporal self-location could be sufficient.

3:AM: What are the implications of your approach to work on AI? There are already examples of machines having representational models – once we start making them with models that represent models, that can learn and so on aren’t we building the rudiments of your ego tunnel – aren’t there real dangers in this?

TM: Absolutely. I have just published a position paper on the issue with a group of young Swiss researchers, unfortunately only in German so far. I have also said for many years that AI and robotics will not get very far without grounded self-models – and that exactly this is also very close to the point at which this research becomes ethically critical. You may know that another side of me is that I have always tried to connect philosophy of mind with applied ethics, and that I have looked into cognitive enhancement, ethical rules for virtual reality, or robot ethics too. I am a vegetarian on ethical grounds for 39 years now, and one interest behind the self-model theory always has been to be able to understand what suffering is, to isolate criteria for what counts as an object of ethical consideration on a more abstract, hardware-independent level. As a consciousness researcher I do not believe that we will have artificial consciousness tomorrow or even on the day after tomorrow – but I may be underestimating synergies and there are smart people already aiming at it today. That is why I think it is important to think about the ethics of synthetic phenomenology right now, in the absence of time pressure and an agitated public, simply because the potential risks are so high.

I have quite a bit to say about this, but, just to put one central point very simply, I would argue for a principle of negative synthetic phenomenology (NSP) stating an ethical norm, which demands that, in artificial systems, we should not aim at the creation or even risk the unexpected emergence of conscious states falling into the phenomenological category of “suffering”. We should avoid increasing the overall amount of involuntary suffering in the universe and not recklessly trigger a second-order evolution before we have understood the deeper structure of our own suffering in much greater depth and detail. We should therefore not deliberately create or even risk the emergence of conscious suffering in artificial or postbiotic agents, unless we have very good reasons to do so. It is interesting to note how the interests of artificial subjects of experience are not effectively represented in any ethics committee or political process today, just like the preferences of future human beings and the very large number of sentient beings that will likely be following us on this planet are only barely taken into account. This has to change. In a nutshell, what we do not want is postbiotic systems that are forced to consciously identify with thwarted or frustrated preferences via a transparent self-model from which they cannot effectively distance themselves.

3:AM: Does meditation help understand the working of the ego tunnel? You are known to practice meditation. Has this been important to your philosophical practice?

TM: Actually this is a very private affair, something that really does not belong into the media, because it concerns my personal life. I would rather share the ten most dramatic failures and all the comical tragedies of my sexual life with you than my meditation experiences! But let me perhaps just point to a few very general issues.
First, if for example you just look at my most recent open-access paper on “M-Autonomy”, which claims that conscious thought actually is a subpersonal process, or my long-standing interest in applied ethics, ideas about the phenomenology of suffering and how it can be minimized, or the whole ongoing research project of understanding the mechanisms by which a self-conscious creature identifies with the content of their self-model, then the influence of meditation practice may be quite obvious. But all that is quite trivial: Every philosopher has his or her own relevance criteria, plus their fundamental theoretical intuitions – and of course they are shaped and conditioned by the phenomenology they have undergone or systematically cultivated in their personal life. A rather uninteresting, contingent fact about myself, isn’t it?

More interesting is the issue of what the notion “understanding” in your question could actually mean. Can there be a philosophically interesting kind of non-discursive knowledge that stands the test of rigorous, analytical epistemology? I have my doubts, because I have been influenced by a more hard-headed context of “There is no knowledge outside of true sentences!” Self-deception is one of the new hot topics in interdisciplinary philosophy of mind, and if there is one best example for theory-contaminated autophenomenological reports, then it might possibly be first-person reports by meditators. The very large majority of serious practitioners of meditation I have met in my life implicitly adhere to some bizarre belief-system or other, many of them are intellectually dishonest in a very fundamental way, which often makes rational communication impossible.

Another question I have been interested in is if there could be what I sometimes call a ”fully secularized spirituality” – or if this is not even a coherent thought. I have never approached the issue on a technical level or in any of my academic publications, but a while ago I have written a popular essay aimed at an interested, general readership and just put it up on my website. The title is “Spirituality and Intellectual Honesty”, and it has had a strong and enduring impact in terms of the reactions from readers outside of academia which I am getting. It seems that many do share my intuition that, in the age of cognitive neuroscience and evolutionary psychology, this has become an absolutely central question in the background: Can a spiritual practice and intellectual honesty be reconciled in a new way?

One can certainly have doubts. Perhaps this is one reason for the interesting fact that there seem to be many more neuroscientists, cognitive scientists, or psychologists who seriously meditate than academic philosophers.
Just as with Indian philosophy, as a personal hobby that didn’t influence my academic work much, I have of course monitored empirical research into meditation and over the years met a number of the leading scientists themselves. I found that many of them try to appear as “secular and rigorous” as possible, but actually are on a mission of some sort. It is certainly true that there is a second wave of fantastic empirical research into meditation going on, and that it is yielding interesting results. But still, my personal impression is that many of the empirical researchers are “crypto-Buddhists” (or something comparable) in their private lives – there often seems to be a hidden agenda. But then again it is only human and natural that people promote and research what they have found to be the most intense and meaningful experiences in their lives. I am not at all interested in doing this, but my sense is that this community could profit from an interaction with philosophers working in theory of science and epistemology – for instance in systematically designing their catalogue of explananda. Meditation research urgently needs the input of good analytic philosophers!

One of the things I find strangest about myself is that I have personally gone quite deeply into what is very dubiously called “first-person methods”, and that I have always found it bizarre and quite suspect how someone could actually claim to do proper philosophy of mind or pretend to have a serious interest in the problem of consciousness without it being understood that they will of course be intensively complementing their theoretical work by exactly such activities. I do not know if you have had the same experience, Richard, but recently I have heard more and more often that something has gone fundamentally wrong with academic philosophy and an increasing number of people have silently begun to ask themselves the question: How does one get philosophy back into philosophy? I think that there may be no royal road and no single silver bullet here, but that exactly a much stronger integration of those dubious “first-person methods” into research and training – by not talking about them, but formally practicing them – would solve a lot of our current problems.

At the same time, I have never really believed in “first-person data” myself, let alone in irreducible “first-person facts”. As a long-term meditator, I am highly critical about philosophical conclusions drawn from contemplative practice itself. There is nothing “given” there. Seriously assuming the existence of such data rests on an extended usage of a concept that is only well-defined in another (namely, scientific) context. “Data” are extracted from the physical world by technical measuring devices, in a public procedure that is well-defined and well-understood, replicable, and improvable; and which is necessarily intersubjective. But in introspecting our own minds we never have any truly direct or immediate access to a mysterious class of “subjective facts”⎯all we have are neural correlates and publicly observable reports (which need not be verbal). To me, all the fancy, romantic talk about “first-person data” rests on an extended usage of a concept that is only well-defined in another context of application, thereby rhetorically exploiting a fallacy of equivocation. “Data” are typically (though not always) gathered with the help of technical measuring devices (and not individual brains) and by groups of people who mutually control and criticize each other’s methods of data-gathering (namely, by large scientific communities). In particular, data are gathered in the context of rational theories aiming at ever better predictions, theories that⎯as opposed to phenomenological reports⎯are open to falsification.

To be sure, autophenomenological reports, theory-contaminated as they may be, are themselves highly valuable and can certainly be treated as data. But the experience “itself” cannot. However, even if one presupposes this rather straightforward view, it would at the same time be unphilosophical to play dumb and deny that all of this cannot be the whole story. Of course there is something relevant there. But before we can take the step from those dubious “first-person methods” to the philosophically much more interesting “zero-person methods” we better develop an empirically grounded and highly differentiated theory of what that vague metaphor of a “first-person perspective” really refers to in the first place – and of what can seriously count as a “method” and what cannot. So, coming back to your original question about meditation and philosophy, perhaps the three most interesting issues are: Can there be a fully secularized form of spirituality, or is this not even a coherent thought? Can we develop an epistemology of contemplative practice (i.e., the philosophically motivated development of non-cognitive and non-intellectual epistemic abilities), or is there really no such thing as a relevant, non-discursive form of knowledge? Thirdly, how would one train a new generation of philosophers of mind and cognitive science who are not only analytically sharp and empirically well-informed, but also “well-traveled” in that they carry out their academic research against the background of a rich experience of unusual states of consciousness and a systematic practice of cultivating their introspective experience?

3:AM: You’re a philosopher who does experiments and works closely with the neuroscientists and cognitive psychologists and other scientists. Is it important that scientists and philosophers learn to work together?

TM: No, I longed to do something unimportant and irrelevant! More seriously, the “learning” you refer to has actually taken place during the last three decades, and now there already is a whole new generation of excellent young researchers who pursue an interdisciplinary approach, worldwide. But it is just so much harder for them! It is much easier to be a really good historician of philosophy and specialize on a specific epoch – if you work hard, then you can build up your expertise and basically be done by the age of 35. The same goes for people who withdraw into analytical metaphysics, or who are very good at formal methods generally: This type of philosophical research is certainly just as respectable and important, but it is much easier to keep at least a general overview. A young philosopher of cognitive science today not only faces the hostility of those still remaining parts of the academic establishment that have slept in on the whole development, but she is also confronted with a never ending flood of new insights and potentially relevant bottom-up constraints from the empirical frontier. If, say, your speciality was “bodily self-consciousness”, “predictive processing”, or “mind-wandering”, then it is hard enough to just stay on top of the philosophical literature. But at the same time you are trying to understand difficult publications from fast-moving neighbouring fields, perhaps with a lot of mathematics, or based on experimental methodologies you do not fully comprehend, simply because you never studied these subjects. The relevant empirical literature is exploding, and many busy neuroscientists may not be interested in more substantial cooperation, because they have acquired a false view of philosophers as those intuition-mongering and slightly unprofessional colleagues who in the end really have nothing more to contribute than Zombie-style thought experiments.

I was always proud to live in a country where all higher education is free and open to everyone. When, however, I spent a year at UC San Diego in 2000 it became quite obvious to me that the philosophy students there – as long as they are able to pay those horrendous fees – not only have a much more attractive environment, much better libraries and green spaces, better air conditioning, food courts and cafeterias, but also that they get a much better academic training than I ever did in Germany. Consequently, I have invested a lot of my personal energy into improving the conditions for German students, for example by helping to initiate the European Platform for Life Sciences, Mind Sciences, and the Humanities, by founding the MIND Group, or by realizing the Open MIND-project together with Jennifer Windt. It took me 6 years to put together a three volume text-book like I would have liked to have one when I embarked on this journey in the beginning 80’s, when almost all there was armchair analytical philosophy of mind.

It remains difficult for early stage researchers today, but in a different way. For a serious young philosopher in this field the current situation is demanding, and highly unsettling. There are very few good and developed models of “empirically informed philosophy”, so many ambitious junior researchers will have to try and develop their very own brand of interdisciplinary philosophy. At the same time there is a shortage of established career paths, there are financial cuttings and overheated competition, plus the fact that the increasing marketization of academic philosophy have also lead to new types of competitors on the philosophical job market: the merciless point-scorer, the shamelessly self-promoting entrepreneur, or the phoney little politician. In a recent poll conducted by the German newspaper DIE ZEIT, 81% of junior researchers and more than half of the assistant professors said they consider leaving academia altogether. Of all the young people I have trained in my career and who have been granted tenure by now, none is in Germany, they all went away – they are in Japan or in Holland, in Canada, Taiwan, or in Australia. We should all keep our fingers crossed for those young philosophers of mind and cognitive science wo are driven by a serious interest – it is by no means an easy world out there.

3:AM: And finally are there five books other than your own that you could recommend to the readers here at 3:AM to help them delve further into your philosophical world?

TM: Sure:
Brentano, F. (1973)[1874]. Psychologie vom empirischen Standpunkt. Erster Band. Hamburg: Meiner.
Churchland, P.M. (1989). A Neurocomputational Perspective. Cambridge, MA: MIT Press.
Clark, A. (2015). Surfing Uncertainty. Oxford: Oxford University Press.
Hohwy, J. (2013). The Predictive Mind. Oxford: Oxford University Press.
Krishnamurti, J. (1976). Krishnamurti’s Notebook. New York: Harper & Row.


Richard Marshall is still biding his time.

Buy his book here to keep him biding!

Pages: 1 2

First published in 3:AM Magazine: Thursday, February 25th, 2016.