Mental states and minds

Towards a typology of mental states

Consciousness

Another word is ubiquitous in discourse about the mind — consciousness. Alas, the polysemy of this umbrella term is formidable and in dire need of being defused before we can go any further.

Let’s consider some of the common meanings of conscious:

Sentience/mindedness: "being conscious of" is often used as a synonym to "being mindful of", so that any mental state would be a state of consciousness. I could be conscious of an experience I’m having, but also conscious of some fact, in the sense that I know that fact. Consciousness and mind thus become essentially synonyms. This is not just down to common parlance: many scientific writings on consciousness are actually about generic mentation activity.
Wakefulness: it is sometimes required that to be conscious, an organism must be awake and normally alert. Experiences in dreams would thus not be considered as conscious.
Phenomenality: "consciousness" is oftenwise used as a complete synonym for "phenomenality", which we’ve outlined above. In a laudable effort to avoid doubts, some authors use "phenomenal consciousness" as their term of choice which, when they entertain no other form of consciousness, can seem strangely pleonastic.
Self-consciousness: the most demanding sense of all might define conscious creatures as those that are not only aware, but also aware that they are aware. A less ambiguous way to describe this ability is as self-consciousness or self-awareness. This meaning has been coming a long way: Descartes already suggested that "to be conscious is both to think and to reflect on one’s thought". This use of "consciousness" as tied to a reflective knowledge about one’s own mental occurences spread rapidly amongst his contemporaries. Locke writes that "consciousness is the perception of what passes in a man’s own mind". In modern terminology, mental states that are about oneself or one’s mental states are called introspective.
Access consciousness: recent authors have embraced this polysemy. Ned Block acknowledges it by calling consciousness a "mongrel concept". Beyond phenomenal consciousness he introduces access-consciouness. Those are states endowed with some representation that can be accessed and broadcast for free use in reasoning, speech or action.

It can’t seem to the reader I’ve accomplished much here. If you were not entirely sure what consciousness was before reading this, you’re surely worse for wear now. Let’s try to be more constructive then, and to make explicit our own understanding of the term. I’ll start from two principles:

All options listed above have some merit. Since the definition of "consciousness" varies so much between authors and eras, since even its use in daily conversations is highly polysemic, it’s for us to take our pick as a convention. There is no logical argument that could lead to an unequivocal choice.
I’ll make the most useful pick. For me, that means a definition that does not make "consciousness" a synonym for one of the other, more unambiguous terms at our disposal: mind, self-awareness, phenomenality, wakefulness,...

So, I don’t want conscious states to be limited to phenomenal states, to introspective (self-conscious) states, nor to wakeful states. That seems to lead to a very inclusive concept, extending to representational states. But I also want to preclude consciousness to be equated with the broader concept of mind. What within the mental world can I contrast conscious states with? The answer is: unconscious states. This is not a humorous answer: the existence of unconscious mental states is actually subject to much polemic in philosophy or psychology.

For our purposes here we will conservatively assume they exist and refine their definition. Unconscious obviously cannot just mean non-conscious, lest we include nearly everything in the universe. A rock, or the Sun are certainly in non-conscious states, but not unconscious. I will call unconscious mental states those that cannot be brought to introspective self-consciousness.

A two-dimensional model

Bringing it all together, and in particular having sharply distinguished between the concepts of conscious and of phenomenal experience, we end up with a two-axes model:

The first dimension is a phenomenal-intentional axis: every mental state exhibits phenomenal and/or intentional properties and can be placed on a spectrum from phenomenal-only (qualia) to intentional-only. In the second part of this section on the mental world, we will focus on the phenomenal-intentional axis and on the relationship between these concepts.
The second dimension would distinguish between self-conscious states (introspection), conscious states (which includes introspection, plus all states that can be accessed for introspection), and unconscious states (hypothetically, mental states that are not available for introspection).

A rough typology could be as follows:

self-conscious phenomenal states (by which I mean "mostly phenomenal", a convention I will keep to): for instance, visual perceptions from an object I am intently looking at, or a throbbing pain in my toe won’t go away.
self-conscious intentional states: that’s any intentional state whose object is my mental state , for example an inner monologue about my own thoughts or feelings.
conscious phenomenal states also include phenomenal experiences that I’m not introspective about, but could be. For instance, a sound that my ears are hearing but has gone "to the background" as I was focusing, or the beating of my heart.
conscious intentional states: any thought, belief, fear, desire, hope, love that is not at the forefront of my stream of thought but could be called up to that level.
unconscious phenomenal states: their existence is controversial. One example that’s often mentioned in the literature is that of blindsight: some patients who have lost sight can still respond to visual stimuli with an accuracy much higher than chance. They deny having any conscious visual experience: whatever visual data the brain is processing cannot be brought to introspection; but the reactivity to the same stimuli suggests we can still call the resulting states phenomenal.
unconscious intentional states: things that "we don’t know we know". Those are closely associated with Freudian theories, where they help fill certain theoretical gaps in the explanation of conscious mental activity and behavior, but also with advances in cognitive science, where Chomsky has proposed the existence of a "universal grammar" that humans are born with and underlies language acquisition, even as it cannot be accessed or put into words.

Disjunctivism and intentionalism

Disjunctivism

Our characterization of mental states has yielded results. I can accept that something is a mental state if and only if it is phenomenal, intentional, or both. Our next question is no less daunting: is there unity in the mental realm? From Descartes and Locke to Brentano and Husserl, the question might have seen specious: mentation was always considered as a single package. But more recent developments have often seemed to study the two marks of the mental on parallel tracks that hardly met. This kind of attitude is sometimes called separatist.

If phenomenality and intentionality have no conceptual connection, there seems to be no good reason to abbreviate "phenomenal or intentional" into "mental". This drains the term of its significance and makes the grouping arbitrary: why not create an adjective to denote "physical and/or phenomenal" states instead?

There are compelling reasons to think there is a conceptual connection: in practice, phenomenal and intentional properties seem to work in lockstep and to be mutually implicating in multiple ways. Setting aside positions on animal cognition, panpsychism or artificial intelligence of a more speculative nature, it appears that one doesn’t go without the other.

That’s not a logical indictment of a disjunctive approach to the phenomenal and the intentional: it has its proponents and is not contradictory per se. But the interface between the two has received increasing attention.

Weak or strong intentionalism

Intentionalism is a general term for theories that propound a specific relationship between phenomenal and intentional (or representational) properties. For this discussion it’s important to remember the three-fold nature of intentional states: they have object, content and mode. In what follows I’m heavily indebted to Chalmers as I’m reproducing the steps of his enlightening discussion of intentionalism.

The simplest possible relation these properties could entertain is that of entailment:

Do phenomenal properties entail intentional properties? This would mean that every phenomenal experience is necessarily accompanied by a certain representation of the world: every pain would be directed at its source, every mood at someone, the world, the self. This position, the simplest step to break with full separatism, is often called weak intentionalism. Following Brentano’s lead, its direct corollary is that all mental states have intentional properties, making the intentional a necessary condition for the mental. A consequence is that pure qualia cannot exist as independent states, yet this version of intentionalism does not seek to explain them; as other phenomenal properties, they are still ontologically autonomous and irreducible. This approach is more characteristic of the "continental" phenomenologist brethren of Brentano, from Husserl to Sartre.
Do intentional properties entail phenomenal properties? That seems hard to argue — as we’ve seen, most intentional states are not self-consciously held. I know that Hugues Capet became the king of France in 987, but before I brought this fact to self-consciousness this instant, that representation did not seem to have phenomenal properties. I can’t say it "feels like something" to know facts about Hugues Capet as compared to not knowing them, the same way it feels like something to have a toothache.

Perhaps we can be more specific then; there is a case to be made that the intentional properties associated with phenomenal experience are characterized by their mode. For instance, knowing I am hurt, seeing my injury, and feeling pain from it would be states with the same object, similar content, and very distinct modes. If we accept that we can circumvent a set of phenomenal modes such as feeling, sensing, seeing, hearing, etc., then we can say by definition that: phenomenal-mode intentional properties entail phenomenal properties.

The combination of "weak intentionalism" with this last assertion is often called strong intentionalism. Since two reverse entailments is an equivalence, this leads to the following corollary: phenomenal properties are equivalent to phenomenal-mode intentional properties.

Reducive or nonreducive intentionalism

Now, that statement may still seem somewhat circular. Once we establish that phenomenal properties are equivalent to some intentional properties, do we need to refer to phenomenality still?

Nonreducive intentionalists do think so. This position can either be a neutral one, for lack of a convincing way to explain phenomenality, or a positive statement that phenomenality is an irreducible feature of the mental, even though it may not be as fundamental as intentionality. Should we stick to this position, the reader may be excused for thinking this is just a big sleigh of hand, replacing "phenomenal properties" with "phenomenal-mode intentional properties". It is not the case: "phenomenal-mode intentional properties" may need to appeal to the phenomenal for their mode, but unlike raw phenomenal properties, have an object and content. Phenomenal properties also supervene on them, in such a way that two beings in the same physical and intentional states would necessarily have the same phenomenal experience.
Reducive intentionalists, on the other hand, believe that the intentional properties that are equivalent to phenomenal properties can be fully understood and described without phenomenal notions. Amongst the numerous ways this has been attempted we can mention functional, causal, teleological approaches, as well as higher-level representation theories. We will not do more than dip our toe here.

Broad or narrow intentionalism

An arduous problem for all forms of intentionalism lies in the question of broad (or wide) and narrow content.

Narrow mental content is entirely determined by a subject’s intrinsic properties.
Broad mental content is not just determined by a subject’s intrinsic properties.

On first encounter it may seem incongruous that we’re even talking of broad content at all. If all things mental are private and subjective, how could mental content depend on anything that’s not solely internal? Sure, we do perceive and feel things from outside our own body, but, following Descartes, the only thing I can be sure of is that these perceptions and feelings are in my mind. In his radically skeptic account, Descartes imagines an "evil deceiver" who could create a perfectly lifelike hallucination or dream state. A more modern example is the proverbial "brain-in-the-vat": if we were able to provide that human brain with the exact physical stimuli that a duplicate "brain-in-a-head" is receiving, many would agree the former would have the same mental states as the latter. It appears to be a clear-cut case that mental content supervenes on intrinsic properties.

But it is important to understand that broad content is not meant to be externally located but externally individuated, which has to do with meaning. This idea shares roots with anti-individualist notions in Sartre, Heidegger or Vygotsky, but it’s been formalized in analytical philosophy by Putnam, whose best-remembered example is that of the Twin-Earths. Let’s imagine there is a twin planet of ours where nearly everything is identical. Every human has a doppelgänger who has lived the same life, gone through the same experiences and has had the same thoughts and beliefs. I’ve said twin-Earth was "nearly identical" to Earth — here is the difference: all the gold on Earth (Au) is replaced on twin-Earth by another element with atomic symbol Xy. It takes the form of a golden-colored, shiny, heavy metal, and its socio-cultural role is exactly the same as gold’s on Earth. Let’s further contend that in twin-England, the word to designate that metal is also "gold". Now let’s consider an English subject from the year 1000, Jon, and his doppelgänger twin-Jon.

Both live in the medieval era, so they have no idea about atomic weights, crystalline structure or fine physical properties. Nothing in their mental make-up would allow them to differentiate between Au and Xy. So, since Jon and twin-Jon have lived the exact same life with the same experiences, on twin planets, with the only actual difference being completely inaccessible to their senses, cognition, and those of any human around them, it seems clear that they areexact duplicates of each other. Therefore they should share all intrinsic properties. And yet, when Jon thinks "gold is shiny" and twin-Jon thinks "gold is shiny", the intentional objects of these thoughts differ: they may not know it, but Jon thinks of Au, and twin-Jon of Xy. Per some definitions of intentional content, the difference in object entails a difference in content. Hence it seems we are indeed facing broad content. Most externalists (believers in broad content) don’t argue that the object itself, Au or Xy, is part of the mental content. Again, it is still internally located. However, the basic idea is that if you fixate the intrinsic properties and variate the external environment, you end up with different mental content.

Externalism v Internalism about mental content is as far from a settled case as it gets in modern philosophy. Both positions have heavyweight proponents and countless arguments and counter-arguments in their favor. One thing to keep in mind is that radical versions of this positions (that there is only broad or only narrow content) stem from different basic intuitions about the mind, based on posits that are arguably impossible to debunk from the other side. Many theories fall towards the middle, accepting that broad and narrow content coexist, or deriving narrow concepts from the content proper.

Let’s bring this discussion back to intentionalism. The gist of this approach is to consider that phenomenal properties are equivalent to certain intentional properties. It seems extremely plausible to say that phenomenal properties are narrow: it’s a very difficult argument to make that the phenomenal experience of pain depends on anything besides one’s nervous system, or that two duplicate individuals with the same eyes, retinas, optical nerves and brains could have different visual experiences. But if phenomenal properties are indeed narrow, then the equivalent intentional properties need to have narrow content as well. This is in contradiction with radical externalism but consistent with radical internalism, and with moderate forms that don’t deny broad content but accept the existence of narrow content.