Talk:I Am a Strange Loop

Capital idea

Whence the capitalization? Is there a style guideline for capitalizing book titles on Wikipedia? I have this book in front of me, but unfortunately the title is listed in all capitals on the dust cover, the half-title, and title page. My own intuition would favor "I am a Strange Loop." Sjeng 19:48, 26 April 2007 (UTC)

Verbs are typically capitalized in title case. - Montréalais 17:51, 4 May 2007 (UTC)

"I Am a" is also Hofstadter's own preference, as you can see on page xv of the Preface. 91.105.24.26 22:07, 13 May 2007 (UTC)

Fair use rationale for Image:Strageloop.jpg

Image:Strageloop.jpg is being used on this article. I notice the image page specifies that the image is being used under fair use but there is no explanation or rationale as to why its use in this Wikipedia article constitutes fair use. In addition to the boilerplate fair use template, you must also write out on the image description page a specific explanation or rationale for why using this image in each article is consistent with fair use.

Please go to the image description page and edit it to include a fair use rationale. Using one of the templates at Wikipedia:Fair use rationale guideline is an easy way to insure that your image is in compliance with Wikipedia policy, but remember that you must complete the template. Do not simply insert a blank template on an image page.

If there is other fair use media, consider checking that you have specified the fair use rationale on the other images used on this page. Note that any fair use images uploaded after 4 May, 2006, and lacking such an explanation will be deleted one week after they have been uploaded, as described on criteria for speedy deletion. If you have any questions please ask them at the Media copyright questions page. Thank you.

BetacommandBot 11:23, 6 July 2007 (UTC)

book reviews

As with many books, amazondotcom is probably the place to start for a wide variety of views of this book -- but I won't bother trying to add such a link to the article, it probably would not last more than a few seconds...-69.87.200.138 00:49, 17 July 2007 (UTC)

yes, it is far more important that this article link to mobius strip and klein bottle, just in case anybody thought to take the book seriously 74.73.179.172 (talk) 18:21, 16 October 2015 (UTC)

I can't quite make out if your remark has an odd or an even number of layers of irony. If odd,I agree that it is not his best - it goes over much of the material of earlier books in a sloppy and free-wheeling style, it contains odd autobiographical digressions, and the nuggets of new insights are disappointing. A core message of the book is a sweeping generalisation of the phenomenon of "statistical mechanical robustness" where the behaviour of epiphenomena is almost wholly divorced from the medium and constituents at a "lower" level and the rules that govern their behaviour. In analogy with stat. mech. we should then feel justified in treating of mental phenomena at a systems level with little or no concern for the neurological substrate; just like classical thermodynamics could (and still can, in principle) be developed without consideration or even knowledge of particle physics. It may be of some interest to readers of this talk page that biological systems offer counterexamples to DRH's sweeping generalisation. The immune system is an exemplar that is relatively easy to establish as such. 2A01:CB0C:CD:D800:5920:D38A:97EF:EAFB (talk) 07:47, 15 February 2023 (UTC)

WikiProject class rating

This article was automatically assessed because at least one WikiProject had rated the article as start, and the rating on other projects was brought up to start class. BetacommandBot 04:08, 10 November 2007 (UTC)

Hofstadter laments that [GEB] was perceived as a hodgepodge of neat things with no central theme.

Fair enough, but why then redo that book worse? 2A01:CB0C:CD:D800:5920:D38A:97EF:EAFB (talk) 08:04, 15 February 2023 (UTC)

"Bertrand Russell did not understand Gödel's work"

DRH makes this claim several times in the present book. Is it true? 2A01:CB0C:CD:D800:4407:8ED8:857:6595 (talk) 16:49, 16 February 2023 (UTC)

"Gödel's name contains the word God"

DRH asserts this in the book, as he also did over and over in interviews following the success of GEB. "And who knows what mysteries might be lurking in those two dots?" - Well, they indicate that the name has alternative spelling Goedel and that the vowel is somewhat like that in spud - unlike the vowel in English God (UK or US pronunciation, take your pick) or even German Gott, which is a bit like that in crop. There is more God in egad than there is in Gödel. 2A01:CB0C:CD:D800:4407:8ED8:857:6595 (talk) 16:54, 16 February 2023 (UTC)

He demonstrates how the properties of self-referential systems, demonstrated most famously in Gödel's incompleteness theorems, can be used to describe the unique properties of minds.

The header of this comment is taken from the article and is a pretty good neutrally phrased summary of what DRH sets out to do in this book. It just leaves open the questions of whether demonstrates is justified ("does he?") and the how and unique beg the question. Thus the section could do with some elaboration.

Ideally we want a crisp delineation of the central thesis of the book, or failing that, a clear definition of "strange loop." I will try to state as clearly as I can what the central ideas are; what follows is lengthy since I have been unable to fit everything together. I have tried to refrain from mixing in my own opinions, except where my best possible précis of DRH's ideas may itself be misleading.

DRH gives two examples and one counterexample to which he returns several times. The examples are Gödel encoding and the human mind (or brain), the counterexample is a CCTV circuit with the camera pointing at the monitor screen (at least I think it is a counterexample; it is the one illustrated on the dust jacket). In Gödel encoding, there are formal statements that are at once isomorphic to expressions of properties of the natural numbers, and also isomorphic to expressions of provability (i.e. derivability) of formal statements within the formal system. In a sense, such a formal system can represent certain claims about itself. Human minds can also do this, as attested by the obvious fact that we do talk and think about ourselves. CCTV watching itself does not do this. So Gödel sentences and humans thinking about themselves are the same sort of self-referential entity, but CCTV is excluded from the party.

DRH does make clear that it is something to do with the richness of internal symbolic representation. A strange loop system can support a very great number of patterns of activity, each of which involving in turn a large number of basic constituents of the system, such that these patterns are triggerable and moreover, stably and systematically and reliably reflect aspects of the world that surrounds the system. But unless the terms in italics are given a more precise and restricted technical sense, as much as what the sentence says is also true of the municipal library down your street (triggering is when somebody takes a book off the shelf, and libraries are nothing if not systematic and reliable). I am not claiming that a library is a strange loop system (i.e. I am not arguing Chinese Rooms to repudiate DRH), I am just saying that the required technical senses elude me and do not appear to be clearly defined in the book. DRH or an expert in his philosophy may be able to clarify.

DRH describes a second (defining?) aspect of strange loop systems: the micro-constituent level is inaccessible to the dynamics at the symbol level. This is true: we do not experience the working of our minds at the level of neurones, synapses, neurotransmitters, limbic circuits, and whatnot. By the same token, a library cannot contain an exhaustive collection of books each of which treats of a particular single glyph in one of them (unless each book is one glyph long, which was ruled out at the outset). It is tricky business to count how many things a person can focus on at any given moment in time, but it is safe to say that it is far fewer than 100 billion. So this second aspect actually seems pretty trivial and, worse, quite irrelevant to the Gödel stuff. DRH talks about creatures on a faraway planet that can only perceive (or interpret) Gödel strings at the systemic level of self-reference. He admits this is fanciful allegory, but it is hard to see what to make of it once we are back down to Earth. There is a loophole (a strange loophole?) which is that repeated recursive definition allows a high-level grasp (manipulation, control) even though relatively few entities interact at the highest level. This leads us to suppose that most of the symbols in the system are not triggered by raw sensory input, but by other symbols in the system (which might suggest a strict hierarchy, but even that much need not be the case). The key property would then not be the inaccessibility of the micro-level, but rather the fact that it is only felt through many layers of recursive definition. Still, if all of this is what DRH means to convey, why is there no discussion of Minsky's Society of Mind? I am not suggesting that the latter is the be-all and end-all on that topic, but if DRH does as I surmise locate the solution in such a partial hierarchy of recursion, why no hat-tip to Minsky... in a book that does take the time to tell us that MM is a personal friend of DRH, and that he invented the word telepresence.

DRH makes a very strong claim about self-perception (or "having an I" i.e. self-awareness): a sufficiently large repertoire of triggerable symbols is all it takes. We all agree that libraries do not become self-aware, no matter how big they become. We are perhaps slightly less sure about the internet, which is more dynamic, responsive, and embedded in the real world, and increasingly autonomic in terms of targeted information retrieval. (To DRH's credit, in the 1970s of GEB he was prescient enough to emphasise the Eliza effect in discussions of the Turing test: the critical question is not whether the computer programme succeeds, but whether the human fails!) But is it just quantity or also quality? If so, how? DRH repeatedly leans toward the side of sheer quantity: a mosquito has a pico-soul, a thermostat has a femto-soul, babies and senile persons have half a soul. Presumably this is all predicated on the respective sizes of the repertoires of triggerable symbols.

But DRH also makes a startlingly qualitative claim: human brains, unlike any other type of brain on our planet (including other mammalian ones), are unlimited in their conceptualising flexibility, i.e. their power to find analogies, that is, to see how one thing is "like" another. What is specifically unlimited here, is the human mind's ability to extend the repertoire of symbols, presumably mostly with "higher-order" symbols. Since there are certainly fewer than 10^(28) atoms in a human brain, and since a symbol in the DRH sense certainly cannot be realised with less than 1 atom, the number of symbols is clearly bounded. Thus DRH must mean our capacity to extend the repertoire in whatever direction is "cognitively adaptive" (to coin a phrase). DRH hedges the claim by stating that there are no obvious bounds on this. DRH gives the example of how a dog will never be able to grasp how telepresence works. The limitations on canine symbol repertoire extension are thus quite apparent. But these limitations are not obvious to a dog; surely our own limitations would not be obvious to us.

Unlimited analogising ability is connected to the ability to form abstractions (also unlimited?) which is somehow related to the ability to form recursive higher-order symbols. Since humans do in fact differ in this ability, or at any rate the fleetness of this ability (which is what we mean when we say that one mathematician is "stronger" than some peer of theirs), DRH must mean that the human brain has an unbounded analogising/abstracting capability in principle. Tempting as this thesis is, it would seem quite difficult to rule out the possibility of insights that lie beyond the innate human power of conception (or rule in, for that matter, which would literally amount to trying to conceive the inconceivable). Ironically, we only have recourse to crude analogy in attempting to think about such superior minds: X is to our mind, as ours is to that of a cat (a stick insect, a bacillus).

DRH makes another claim, which is that the physical nature of the micro-constituent is immaterial as far as the level of the symbols is concerned (or rather the level immediately above that, where we describe the dynamics of how symbols interact). Biology is replete with examples in which systems-level properties only make sense in the light of molecular/cellular properties, and vice versa. DRH dismisses these as very exceptional. This would seem to be premature, given that the only full-fledged physical example of a strange loop is that of the human brain, which is a biological system.

Finally, DRH claims that all or some of person X's repertoire of triggerable symbols that stably and systematically and reliably reflect the world may also occur in person Y's mind. Human communication presupposes some correlation between different people's internal representations, and what else is daily existence but a continual confirmation that this correlation is far from perfect? But DRH goes much further and suggest that a largish core portion of person X may, in effect, be transplanted to the mind of person Y, as attested by how "in synch" people who spend much time together can be. DRH suggests that this can serve as a source of consolation after a loved one dies. DRH reasons as follows: things are real to us inasmuch as they occur as active symbols in our minds; this goes for people as well as for any other kind of entity; the self is just the people/person concept/symbol for the special case of that individual who inhabits the brain (and is charged with guiding it and its attached body through the world); therefore the people concepts/symbols for the general case of everybody else is just how their other selves inhabit the brain of ego. Admittedly, the self/ego "I" symbol is very much richer than any other representation we might have of other people in our brains, but to DRH this is only a difference in degree, not in nature. (Incidentally, we don't just have internal representations of other people, but also representations - in ourselves - of how we are represented in them; infinite regress threatens, but de facto this is the recursion level where we stop.) DRH says that the representations in our brains of other people are low-resolution, coarse-grained versions of the selves that inhabit their own brains. (Again a quasi-editorial aside to forestall misunderstandings: these representations are not equally coarse-grained across the board. We can predict other people's reactions to certain situations with very high fidelity and draw a complete blank otherwise. You can have lunch with your friend the heart surgeon every day for 40 years and while you predict quite accurately what she will be saying tomorrow about the latest political developments, you have no clue how she will deal with tomorrow's eventualities in the cardiac operating theatre. I am sorry if this is a blatant truism, but DRH systematically underplays just how fragmentary our insight into other people's inner lives is.)

2A01:CB0C:CD:D800:1859:B404:DC8:1423 (talk) 09:11, 21 February 2023 (UTC)