He actually explains this in the preface to Philosophical Investigations:

For since beginning to occupy myself with philosophy again, sixteen years ago, I have been forced to recognize grave mistakes in what I wrote in that first book. I was helped to realize these mistakes - to a degree which I myself am hardly able to estimate - by the criticism which my idea encountered from Frank Ramsey, with whom I discussed them in numerable conversations during the last two years of his life. Even more than to this – always certain and forcible – criticism I am indebted to that which a teacher of this university, Mr. P. Sraffa, for many years unceasingly practised on my thoughts. I am indebted to this stimulus for the most consequential ideas of this book. — Wittgenstein

He came to see the Tractatus writer as playing one among many possible language games. That's the simplified version of how I read it. That doesn't stop the Tractatus being a kind of summation of a certain point of view, a pov which can still feel relevant and useful. He discussed with Nicholas Bachtin publishing the two books together; I think this shows he still valued the earlier work. Still, the P I for me has great depths including challenges to critique most propositions like yours or mine.

For example, the success of computers in modeling reality via computational means (mathematical truths and logical formulas) in my eyes validates the findings of the Tractatus over the Investigations. — Question

I'm really not sure how computers doing what computers are good at validates any observations on human thinking and communication. Of course there are elements of human thought that bear some resemblance to computation but it is no more than a part of the extremely complex whole and, largely consisting of autonomous and unconscious functions, a small part at that..

The best way to understand is to study the "middle Wittgenstein" to show the connections between the Tractatus and Philosophical Investigations project, which I think is missing from a lot of people's Wittgenstein studies and I think is quite awful.

Philosophical Remarks and Big Typescript are the first recommended works that pretty much outlines in Wittgenstein's words what he was trying to get at in Philosophical Investigations based on his earlier ideas in the Tractatus.

There are plenty of other notebooks and lectures from Wittgenstein that were found and published that helps.

Computers are, currently at least, the best metaphor we have for the workings of the human brain. So it isn't unreasonable to extrapolate from computers to hypothesize that the human brain must work in similar ways.

Apart from the descriptive stance on the difference between the Tractatus and Investigations, I find the Tractatus one of the most aesthetically pleasing works of art that I have encountered in my studies of philosophy.

Everything in it follows seamlessly and effortlessly from one proposition to the next. Only a man of such genius as Wittgenstein could have written such a holistically pleasing book about the correspondence between language and reality.

I'm wondering if anyone else shares this view.

Essentially one can conclude that ethics (one of the main 'unspoken' points about the Tractatus) is either according to the Tractatus is a synthetic a priori endeavor, whereas less intellectually pleasing according to the Investigations is a culturally normative or simply a type of language game derived from cultural norms.

I should state that I'm pretty much a Platonist in my views and the Tractatus hits pretty hard home in that regard.

This is a very good point brought up. I have posted a little about this in the past and subscribe to the notion of a "logical space". As a Platonist I hold the view that reality can be described by mathematics and logical truths. Obviously Godel's Incompleteness Theorem is a snag in presenting reality in terms of logical truths alone, and, have wondered what would the early Wittgenstein would have said about Godel's Incompleteness theorem. Any input appreciated.

You do realise that you've just presented an entirely circular argument? Computers are the best analogue of human brains so we figure that human brains must work something like computers? How do we judge that computers are the best analogue? Cos they work a bit like human brains! Needs some work does that!

It is all, of course, irrelevant to Wittgenstein, anyway, who had no experience of computers and nothing but actual human behaviour to work with but I really can't understand this love of the computer analogy. It is on every conceivable level a load of dingo's kidneys, frankly! Just compare the completely automatic way that we catch a ball, say, to the mass of range finding and calculation and recalculation and signalling of moving parts that a robot would require. When you come up with a computer that can recognise and immediately respond to a joke that nobody's ever heard before then I might concede that a computer is thinking more like a human (though personally I do not expect that day to ever dawn). That's as far as I will ever go.

How do we judge that computers are the best analogue? — Barry Etheridge

Turing test 'em. And as far as I know, AI is being modeled based on... human intelligence.

I sincerely hope that that is not the case. There is no advantage to having AIs that are wrong as much as they're right, prone to abandoning logic and rationality for emotion, and avoid thinking altogether whenever an 'obvious' answer (usually wrong!) presents itself!

You do realise that you've just presented an entirely circular argument? Computers are the best analogue of human brains so we figure that human brains must work something like computers? How do we judge that computers are the best analogue? Cos they work a bit like human brains! Needs some work does that! — Barry Etheridge

I believe you have failed to comprehend the argument. So here's how it goes, explained in a simpler fashion.

We first observe human brains and their functions and list down the properties we observe.

Afterwards, we observe computing machines and the way they function and list down their properties separately and independent of the previous list.

Now we compare the two lists and note that they share many similarities (eg: both human brains and computers are used for the storage and processing of information), while also noting that their differences are comparatively very few.

We now consider other candidate entities which we suspect to be analogous to human brains and put them through the exact same process outlined above.

Now you would see that computers are the highest on the list of potential candidates in the aspect of sharing similarities with the human brain.

Thus, we conclude that computers are the best analogue for the human brain.

Pray explain how that is circular?

I'm not sure what Wittgenstein would have said, but I think "logical truths'' ,in the binary true/false sense of the term, was disposed of anyway in explaining reality, almost immediately after the discovery of quantum physics.

I know nothing much about this sort of thing myself, but I've seen somewhere that quantum physics implies that it is possible for a quantum particle to be in two places at the same time.

Besides what is a ''logical space''?

That isn't (though your previous expression of it was). It's just wrong. A list of properties of the human brain bears almost no points of similarity with the properties of a computer. The ability to come up with the same answers to a limited set of problems does not in any way provide evidence of similarity of process. And it's extremely disingenuous to say that computers process information unless you're prepared to accept that binary code is an adequate definition thereof. Computers simply do not think. They have absolutely no awareness of what is passing through their logic gates and no ability to differentiate. They have no capacity for doubt, for self-correction, for originality or insight. And most significantly of all they are not self-initiating. Anything a computer does that looks even remotely human it does at the behest of a human programmer. They are simply calculating machines, no more than sophisticated abacuses in the final analysis. Are you really suggesting that an abacus is a close analogue for the human brain?

it is possible for a quantum particle to be in two places at the same time. — hunterkf5732

Oh, it's far more complicated and weird than that!

I never said that computers bear a fantastic, twin-like resemblance to the human brain.

My contention all along, was that computers are better analogues than any other entity (and hence the best analogue) for the functions of the human brain.

I see nothing in your reply that disproves this claim.

Computers simply do not think. They have absolutely no awareness of what is passing through their logic gates and no ability to differentiate. They have no capacity for doubt, for self-correction, for originality or insight. And most significantly of all they are not self-initiating. Anything a computer does that looks even remotely human it does at the behest of a human programmer. — Barry Etheridge

Listing a whole lot of differences between a computer and a human brain doesn't disprove the claim either.

It's not a whole lot of differences. It's a list of examples of why pretty much everybody in neuroscience agrees that computers are not an imperfect analogue but no analogue at all.

both human brains and computers are used for the storage and processing of information — hunterkf5732

You say that computers are not an analogue at all to human brains.

Thus, a single similarity would suffice, to show that at least in this sense, they are analogous to human brains.

Besides what is a ''logical space''? — hunterkf5732

I believe that question was answered at PF, while it was still around.

Logical space.

I think a better description of logical space can be given. Basically, using computational understanding the logical space of this forum enables us to communicate effectively with one another. Thus, this logical space is consistent and the representation of it is defined to be the algorithms working effectively to represent this forum. Facts are what things that allow us to communicate effectively on this forum, eg. algorithms, logical truths, math, etc.

So, as a Platonist one can say that the logical space of the universe is defined as the sum total of physical laws in combination with mathematical truths.

Furthermore, if all the logical truths, mathematical truths, and physical laws can be computed, then this would further support the notion that the logical space of the world are consistent and true (working).