Why is the concept of self-awareness important?

Question

When we discuss whether something (an animal, a strong AI, a Boltzmann brain, etc) would have the kind of intelligence like a human, there are often materials saying whether they would be possibly self-aware.

Now when we write a program for peer to peer networks, simple unintelligent game AIs, etc, it's not unlikely we will model itself like every other peer or fraction. And it will analyze its own attributes like every other peer or fraction, possibly with some extensions. Yet we would not likely link this to the concept of self-awareness. We could say they aren't capable of sufficiently understanding themselves. But, many if not most of the humans are also unable to completely understand themselves. If a simple program is not too different from a human, sure it shouldn't be so significant. But where to draw the line?

In short, self-awareness in the lowest standard is trivially easy to obtain, in the highest standard it is not obtained by most humans. What's the "self-awareness" that sounds like would distinguish humans from other things in a meaningful manner?

About the potential duplicate: This question isn't actually relevant to the other concepts. I'm only asking about the significance of self-awareness. Mixing up with other concepts is one possibility that implies insignificance in some situations, and could possibly explain why I felt it is significant. But it isn't relevant if it is actually significant. I have removed the paragraph that mentioned the other concepts to reduce confusions.

Answer 1

Because it deals with the nature of assumption. When we are aware, we assume reality.

Thus self awareness is the assumption of assumption, where one self evident truth (assumed) projects to a further self evident truth (assumed).

This had implications in not only consciousness but also the maths, sciences, religions etc. precisely because they are all built on axioms...which are assumed.

Even conceptualizing a phenomenon, ie to assume it from another angle, results in a change in that phenomenon.

A simple example would be the the principle against circularity by Peter D. Klein under the infinitism school. It is a principle that necessitates one axiom progresses to another but it never repeated. So one assumption effectively changes to another non repeated assumption and this provides a definition of "how to reason" without necessarily "self reflect"...which is circular.

However significance is contextual...again.

It is significant for AI of course, but is necessary for quantum physics in light of the paradoxes of "measuring something affects it" which you can Google and get a load of interpretations...which goes to basic questions of how we derive self evident truths all together in the scientific method. You can read popper for the question of science.

Also googling manly P hall, point line and circle will give some clarity as to the nature, hence importance of self reflection.

Basic self reflection begins with apriori concepts (read further into kant). The most basic is the point. (M.P. Hall).

Considering we observe through apriori concepts we effectively reason through them. Space and time is the most subject of them all (kant, Augustine).

So in assuming oneself we effectively are assuming a point of awareness. This is represented by a simple point. In assuming this state of awareness we create a new self, a new perspectivd... quite literally a new point of awareness.

This can be observed in our basic conception of time, the grounding for subjectivity, as represented by a line between to points (as observed by Heidegger in being and Time).

One subjective self, with the most subjective assumption being a simple point, in reflecting upon itself effectively individuated itself into another point of awareness.

This is the foundation of definition...one point directed to another point with each point not only representing a self as point fo awareness but also a measurement in empirical reality where we pick a point of reality and measure from it. Euclid observed this in his first postulate but never elaborated much on the psychology of it.

The pyrhonists observed this inherent dissent of one point of view to another, in agrippas trillema.

So the dividing line whic literally is the progress of one set of assumed truth to another and we are left with the many different school and sciences...and further confusion.

Platonic forms, specifically the line between two points, can be observed as the foundation of self reflection as having an inherent form. Considering Plato himself set up the argument of Justifed true belief, one could argue that the forms are strictly "the way we reason".

This can be observed in the forms of linear and circular reason we use to define any assumption we make as well as this assumption merely being a "point of view" in itself.

Hopefully this will give some food for thought, I may have to clarify some points further.

But the paradox is that self reflection is the act of definition.

Answer 2

I will try to give what I think is Sebastian Rödl’s answer to this question. Here is the very first paragraph of his book Self-Consciousness (p. vii):

The topic of this book is self-consciousness. Its chapters treat of action, of belief, of reason and freedom as a material reality, of receptive knowledge, and of the concept of the second person. Of course, each of these topics deserves its own book. And yet these books, whether they acknowledge it or not, will all be books on self-consciousness, for self-consciousness is the principle of their subject matter.

In the second paragraph of «The First Person and Self-Knowledge in Analytic Philosophy» (pp. 280–281) he makes an important distinction between two ways of understanding the role of «self» in these words. Applying that distinction here is relevant: Self-consciousness, self-knowledge, or self-awareness is important because (and in so far as) the prefix «self» does not signify a special object of consciousness, knowledge or awareness («the self»), but because (and in so far as) it marks the form of human consciousness, knowledge or awareness.

The very idea of the form of knowledge is difficult, but this claim (that the form of human knowledge is its self-consciousness) comes from Kant. It can be found in his famous statement (from The Critique of Pure Reason) that it must be possible to attach «I think» to all my representations (B131). Any representation is self-consciousness in this sense, no matter what its object is. The point, as it applies to knowledge, could be put like this: Being in a position to truthfully say and mean «I know that the dishwasher is broken» is nothing other than being in a position to truthfully say and mean «The dishwasher is broken». The «I know» comes «for free» because it is just an explication of the self-knowledge already in «The dishwasher is broken».

So one answer to your question is that this formal feature of human knowledge is important because it is what defines us.

Answer 3

I think that to some degree, the focus on self-anything is overstated in our philosophical tradition. So I would avoid using self-awareness as a sort of definition of human-like thought.

Daniel Dennet (in Consciousness, Explained, and elsewhere before that) points out that since the brain is 'multithreaded', we are always concurrently many selves. We may only really have experiences tied to the single self, like self-awareness, in conscious memory, which happens only on reflection, when we encode our experience in language and symbolic markers. It never happens in real time. Self-awareness, then, may only exist to its fullest degree when we are surrounded by others and dependent upon them. Because of our sociability, we seem to habitually encode our experience in a communicable form. We may do so only because of the importance of others to us, and not for any more essential reason. The intelligence that creates self-awareness, then, would not necessarily be changed if this circumstance changed, and the habit went with it, but the kind of self-awareness might be completely alien to our ordinary picture of it.

From a psychological point of view, what is relevant to intelligence is the ability to make full use of the environment. In an environment containing animals, this involves developing 'theory of mind', the ability to see other beings as agents rather than things, so that you can use them as allies or defend your claims against theirs rather than seeing them only as tools and materials that are not cooperating or competing with you. Otherwise, a large part of the power available in the environment remains untapped.

This has self-awareness as a corollary. The ability to make use of the agency of others to predict behavior implies the ability to see it in yourself. To the degree, for instance, a herding dog sees agency in its master, and to a lesser degree in the sheep, and can use that understanding of agency productively, it is much more intelligent than the sheep, whose main picture of the world is largely as a supply of grass with embedded threats. The sheep cannot use their understanding of the agency of wolves to their own advantage -- they are of a lesser order of intelligence.

Humans of average and above intelligence have very complete theory of mind relative to other humans, they can attain true empathy, when doing so makes sense, understanding one another to a degree that allows for disagreements to be worked out so that immense social groups can function and still leave individuals with substantial agency.

The sheep, like schooling fish or even ants or bees understand each other's behavior, but through actual similarity, and not through projective modeling. Each sheep or ant can predict the behavior of others because their individual agency is delimited to such a degree that as individuals they can in fact not productively use the environment for their own ends. The group has a degree of intelligence greater than the sum of the individuals'.

For humans, often the group is of a lower order of intelligence than the individuals in it. We do use this same kind of communal intelligence, but when individuals sacrifice agency to a common purpose and align themselves as a herd, the whole will contain individuals or smaller groups who are more intelligent than the entire group acting in concert, and they will be able to guide the entire mass for their own purposes.

But then self-awareness is not the point. Leverage of the theory of mind is the point, and an AI that could successfully manipulate the behavior of others consistently would have met the target. It would, as a side-effect, have some form of self-awareness, but it may not be one that we would readily identify.

Answer 4

Self-awareness is necessary for self-change; it constitutes the distinction between a 'subject' and an 'object'. Objects — things that lack self-awareness — are incapable of change except as acted on by some process or force. They are subject to inertia (in the expanded Newtonian sense) and do nothing except by the dictates of natural laws. *Subjects° — things that have self-awareness — are ostensibly capable of modifying themselves (acting as a force on the object that is themselves). Peer to peer networks are conditional but not self-aware; they have a finite assortment of switches that produce different behavior in different environmental contexts, but they are unaware that those switches exist and cannot modify them at will.

I trust you can see why philosophy might think this is an important concern. The fact that most humans do not seem to engage in it is irrelevant; the question is whether humans are capable of self-awareness, because that opens the entire dimension of moral behavior for consideration and analysis.