If the mind is required or constituitive for beliefs, statements, and perceptions, then it is just too selective and misleading to claim that the mind is representative. Perceptions are not representative, and therefore there is a part of the mind which is not representative. The experience of an object that one perceives does not represent or symbolize some other non-perceptible object; the experience is the presentation of the object that you perceive.

I would suggest the homunculus problem. To what are representations presented? Any theory of mind is a representation of mind to itself, and so must necessarily be partial at best.

The experience of an object that one perceives does not represent or symbolize some other non-perceptible object; the experience is the presentation of the object that you perceive. — jkop

But it is followed immediately by a perceptual judgment, the involuntary (i.e., acritical) representation of the percept in thought. Much like a retroductive conjecture, this then serves as the first premiss for any further reasoning about the percept, since the percept itself is gone as soon as it is experienced.

perceptual judgment, the involuntary (i.e., acritical) representation of the percept in thought — aletheist

It is not your judgements that you perceive but objects and states of affairs. Otherwise you would never see the latter, only your own figments of mind, which could then only "represent" eachother.

Right - you perceive (non-representationally) the objects and states of affairs, and then you judge (representationally) what you have perceived. In other words, perception is presentation, and any subsequent thought about it is representation, since the original percept itself is no longer present.

since the original percept itself is no longer present — aletheist

Hold it. Did you not see objects and states of affairs after all but some percept that disappears?

I perceived objects and states of affairs, which produced a percept, which disappeared immediately. The perceptual judgment that followed is what I have been thinking about subsequently. In that sense, the actual objects and states of affairs that I perceived constitute the real object of those thoughts - i.e., representations.

Davidson said that Humean impressions are not epistemological contents. This follows that perceptual contents need linguistic contents to be epistemological ones.

Once you assume that perception would produce a percept or sense-datum the object of perception will certainly seem representational, just like a thought, belief or statement. Hence the need to quickly sweep the percept under the rug, it simply disappears and is replaced by a thought.

But the assumption is false, there exists no such thing as a percept, the idea is completely ad-hoc. See, for exmple, Austin's Sense and Sensibilia (1962) in which he pulls apart various sense-data theories. We can also compare some of the properties that set objects of perception apart from representative thoughts.

Unlike thoughts and beliefs about objects, perceived objects typically change in a continuous flow as they are being perceived. They're changing in the here and now, and the perceiver can't simply detach what s/he perceives, nor reinterpret it by will power. Moreover, objects of perception don't have the disjoint syntax of language, such as in statements or descriptions.

It seems fairly clear to me that perception is one counter example to the claim that the mind would be representational.

A humean impression is the old version of sense-data. But sense-data is ad-hoc, it does not really exist.

Indirect realists say that what we perceive are sense-data of objects. Direct realists say that what we perceive are real objects themselves.
This follows that you are a direct realist, aren't you?

Yes. In the vein of Searle, who was a student of Austin.

Are we disagreeing? We (involuntarily) perceive the objects as they are presented to us, then we think about representations of them.

I thought you might appreciate this vid.

Is there a written version? I am a reader, not a watcher. I did enjoy Kahneman's book (with Tversky), Thinking, Fast and Slow.

One of my objections would be this: the only way that a Turing machine manipulating symbols in a purely syntactical manner and executing truth-preserving algorithims "syntactically" works in the first place is via us thinking about/imagining such a thing and thinking about it with semantic associations (perhaps without someone fully being aware that they're applying semantic content). So saying that our brains could work analogously, so that there is a "level" wherein syntactical manipulations executing alogrithims preserve truth for something like a modus ponens, doesn't hold water. There's no reason to believe that such a thing is even coherent, in any arena.

Re comments in the thread in general, Fodor's representational theory of mind is not a theory of perception. It's a theory of how mind works in general. "Represenational" is an unfortunate word for it, because of the well-known reprepresentational approach to philosophy of perception. Fodor's view is more of a "symbolic theory of mind" in the sense of a "symbol-manipulation theory of mind."

I don't think the vid covers much new ground if you have read thinking fast and slow.
I have not read the book myself, but in the vid he does mention fast and slow thinking.
He also encourages the view that there are separate but distinct ways in which people use the term know.
One in which people are expressing that they do not doubt something.
Then there is the way that science employs the term which is to indicate there is strong evidence and valid logic to accept the truth of something.
He goes further to explain the different ways in which we associate our immediate experience with knowledge, where at that point he covers the notion of fast and slow thinking.