[The start of reference 2]
Contents
• Introduction
• Other approaches
• What I believe
• Some other points of detail arising.
• Other matters
• Conclusions
• References
Introduction
I came across a printed copy of reference 1 just before we went on holiday, tucked into something or other. I thought that I must have printed it off in the hope that I would get around to reading it one day, so, armed with this thought, I took it on holiday with me and I did try to read it. But I soon moved on to reference 2, which I found a bit more accessible.
Notwithstanding spending quality time on the latter, I have failed to understand what Davidson, an eminent philosopher of the last century, was on about, despite having a reasonable background in both language and logic. I failed to compute much of the detail of what he was saying, never mind the shape of his argument. Or even what his argument achieved, where it got us.
This post tries to set down where I got to. Interested readers can compare the present post with last year’s, at reference 3.
The subject matter of both these papers is the difference between what I know of my own mind, from the inside, as it were, and what I know of the mind of others, very much from the outside. What I know of the minds of others, I derive from their behaviour, what they say and what they do. It seems self-evident that I know my own mind better than those of others, but is that enough? Is there more that can usefully be said?
Davidson clearly thinks that the answer to these questions is no and yes respectively – but he does not think that the analogy of an eye-witness is helpful; the idea that I am an eye-witness of what is going on in my own mind, but I only have hearsay about what is going on in that of others. While I thought that it was quite a good analogy.
Still and all, very much related to what is called elsewhere the theory of mind, all about children learning that other people have minds rather like their own, but minds which hold rather different beliefs and desires than their own. Eventually they learn how to navigate those other beliefs and desires, an important part of getting along in the world. It is not enough to know what colour that brick is; I also need to know what colour you think it is.
Davidson restricts his attention to a restricted class of thoughts expressing first person, present tense, propositional attitudes, which last are explained at references 8, 9 and 10. Roughly speaking, thoughts or utterances of the form ‘I believe that this book is too long’, in which the belief is the attitude and the book is the subject of the proposition. Or, more tricky: ‘Jill believed that Jack ran up the hill’. Tricky because this one is neither first person nor present tense. Two points here.
First, I do not see propositional attitudes being a big part of what I think or say. Maybe they are there in the background, maybe they are important building bricks of what I know and feel about things, but I do not spend quality time rehearsing them, at least, not in that explicit way.
Second, I do not see why Davidson’s makes this restriction, but maybe it is mixed up with the observation at the beginning of reference 10 that:
‘… Having a successful theory of propositional attitude reports is important, as they serve as a converging point for a number of different fields, including philosophy of language, natural language semantics, philosophy of mind, metaphysics, and epistemology…’.
And it is certainly clear from reference 10 that these propositions and attitudes are more tricky than might at first appear.
One problem of both papers was the amount of jargon of just this sort, the arcane jargon of the professional philosopher talking to other professionals. Jargon which includes using ordinary words with special meanings, related to but not at all the same as their regular meanings. A problem mitigated by access to the Internet and Wikipedia, with some of this mitigation being reflected in the secondary references below. But even so, it does slow one down.
A more serious problem is that I get the impression that Davidson and his colleagues are thrashing about in empty space. They are not properly connected to the real world, at least the real world as it seems to me, forty years on from the writing of these two papers. I associate to the angels of reference 4. Also to the post at reference 5: the gulf between what I know of my own mind and what I know of those of others is not as big as it once was.
The philosophers talk about thoughts without giving proper space to the neural processes which give rise to thoughts. To my mind, trying to come to interesting conclusions about these matters by cerebrating in vacuo, as it were, is the wrong approach. A much better approach is to try and find out what thoughts and beliefs amounted to in neural terms, to leverage the extensive science hard science of mind that is now available.
It is all very well to cerebrate about, for example, Abelian groups, the complex plane or even the deity, in vacuo, because, oxymoronically, that is where they are to be found. But the brain is not like that; it is a complex piece of machinery which can be observed and studied and, when appropriate, taken to pieces. It is chemistry, physics and systems, not in vacuo at all.
Other approaches
More than two thousand years ago, a chap in India thought he could write down all that one needed to know about the pronunciation and grammar of the Sanskrit of his day in around 4,000 verses, aka sutras, in that same language. The first ten of them being snapped above – as turned up by Google AI. The word count is around 8,000, maybe 80 printed pages – without commentary or apparatus – and it can be recited in a couple of hours. It was, after all, part of a largely oral tradition – and one of the drivers was a desire to preserve the chants in temples, unchanged for all time. A bit like the Latin of the Roman church – and both Sanskrit and Latin ended up as dead, priestly languages. With as, it happens, Latin being a relative of Sanskrit, with both being in the Indo-Hittite family of languages. See, for example, reference 19.
By all accounts, an extraordinary feat of compression, with one point of interest being that you were defining something in terms of itself. Some description of this definition is to be found at reference 12.
Since then, there have been lots of attempts to describe natural languages in more or less formal terms – maybe even logical. Describing computer languages was a lot easier, not least because the description tended to come before the language, rather than the other way around. While early in the last century we had Whitehead and Russell’s ‘Principia Mathematica’.
The present paper in very much in this tradition. While forty years on, Gemini and his friends have been remarkably successful without, to my knowledge, bothering with this sort of thing at all; more statistical than analytical.
That said, the rules must be in there somewhere, and it would be interesting to explore how exactly the rules of grammar get expressed inside systems like Gemini.
In any event, my intuition is that, as things have turned out, Davidson and his friends were barking up the wrong tree, interesting tree though it may well be.
What I believe
The starting point of reference 1 was promising. That there was something fundamentally different about what I know about what I am thinking, my own present tense beliefs, from what I believe, or think that I know, about what other people are thinking.
Until recently at least, this last was based on what they did and said and this was not necessarily the same as what they believed. I might say things that I do not believe for all kinds of reasons, some good, some bad, but I still know what I do believe, whatever I might say. And it seems reasonable to suppose that, in this matter at least, other people are much the same as I.
Maybe ‘believe’, not in the sense that I spend quality time rehearsing belief statements as inner thought, rather in the sense of what I might think if someone or something prompted me to think about something. What I might say, however, is another matter. If, for example, the host at a dinner party asked what I thought of his roast beef; when I stopped to think, I might be clear enough in my own mind that I did not like it – perhaps I did not believe that it had been properly cooked, cooked to my taste – but I might well be unlikely to say so. An utterance is a performance, a social act, and my attitude to the proposition at hand is only one ingredient of that performance.
Or am I clear enough in my own mind? Sometimes I deceive myself, I maintain a belief which does not have the support of evidence, the evidence available to me. Think of being in denial about having an unpleasant illness, or something unpleasant coming off in one’s personal or public life. For example, the uttered word ‘prostrate’ really can have power over elderly men. Or perhaps about the performance of some shares which one has hung onto, perhaps for sentimental reasons, perhaps simply because one can’t be bothered to do anything about it. Maybe what I am saying here is that the proposition can be straightforward enough, but my attitude to it might still be complicated o messy.
Another possibility is that someone else who knows me well enough, knows which way I will jump about something when pushed, what I will do or say if pushed into a corner on the matter. In this sense, he would know my mind better than I do.
A less dramatic example would be the bored staff of a shop or café looking out at possible customers peering in, trying to make their minds up whether or not to come in. They make bets about it and some of them are quite good at it. In this restricted sense, they know the minds of these possible customers better than these last know themselves.
Then when I used to play cards, say cribbage, some of the people that I used to play with had an uncanny knowledge of what I was going to do – with the result that I usually lost. Again, they knew my mind better than I did.
A rather different complication is the use of the word ‘believe’. When I say to someone ‘I believe that the plant growing up outside the door of our holiday cottage here at Brading is a confederate jasmine’, my belief does include the possibility that I might be wrong. I am open to correction. I would be being more definite if I were to leave out the qualifying ‘I believe that’. So for me, ‘believe’ is a fairly weak assertion. For the jasmine see reference 7.
Rather different again, in writing these words, there is the thought in the background, a thought that I have had often enough before, that the act of keying stuff into the laptop is what precipitates the thought into the public domain. Draws up the primordial soup from the brain and shapes it into something interesting, useful, or at least coherent. Channels that soup, gives it direction, force and movement. The soup has been canalised.
Maybe something of the same sort sometimes goes on in the heads of good public speakers. They have thought about what they are going to say beforehand, maybe formed a rough outline – but they have not written out their speech and they do not use notes. They just get up on their feet and get going. The force is with them. I seem to remember that Winston Churchill could do something like this, until one day he dried up, and after that, certainly for big set piece speeches, he had the whole thing written down, together with pauses, emphasis and all the rest of it; cunningly crafted to sound both spontaneous and deeply felt. I associate to the time it takes to get the hang of using an autocue on camera.
Or maybe conscious thinking is a more layered business. The unconscious brain might generate all kinds of stuff, have all kinds of thoughts, some of which are nonsense and some of which are unpleasant. Most of this is filtered out before it reaches consciousness, at least when we are not dreaming. Notwithstanding, a fair proportion of what reaches consciousness is still nonsense; a proportion which is dismissed by the concentrated processing applied to (the single threaded) conscious thoughts – and which do not make it to speech.
From where I associate to the super-ego of the Freudians, another kind of supervisor. Also to the way that these supervisory functions weaken with age, with inner thoughts making it into speech without much check on the way up from the inner depths. Also to the way that there is not much room for inner thoughts or checking at all when an orator orates: the words just flow and, again thinking with my fingers, it seems surprising today that so little rubbish slips out.
Which is just as well, as in some contexts, my word is my bond. I will do or believe what I have said. I flip to really believing what I have said in the heat of the moment.
On the other hand, if I were more devious, more calculating, I might well say stuff which I know that I do not really believe, just to test the water, to see what others are thinking.
Maybe there is room for some of Hurlburt’s Descriptive Experience Sampling here, which I poked around a bit, near ten years ago now. See, for example, reference 18.
So the relationships between what I believe, what I think, what I say and what you make of what I say are all very complicated, all very interesting. But am I any further forward?
Some other points of detail arising
Page numbers, where given, refer to reference 2.
At page 102, I got to thinking about when I had thoughts about things, thoughts which might count as propositions which I believed. So as I type this, I am looking around the room, vaguely eying the various objects around me. But, generally speaking, I am not having conscious thoughts about those objects. I am not saying to myself, for example, that this object is a jug and that object is a toaster. Let alone ‘I believe that this object is a jug’ or ‘I believe that that object is a toaster’. Never mind ‘BH thinks that that object is a toaster’. Just occasionally, perhaps when I do not at first work out what something is, its common name, for example ‘toaster’, might pop into consciousness. Or even: ‘the toaster that I picked up years ago at a car boot sale in Hook Road Arena’. But more often, I will just pass on, without conscious comment.
Moving on, an important feature of propositional attitudes is that the proposition can itself be a propositional attitude. I call this circularity and I dare say it is related to what mathematicians and logicians call recursion. This feature means that propositional attitudes can, in theory, get quite deep:
Jack believes that Jill thinks that Fred is a bit of a twat.
They can also get quite silly:
Alpha believes that Beta believes that Charlie believes that Delta believes that Echo believes that Foxtrot is a sort of omelette.
Or worse still:
Alpha hopes that Beta believes that Charlie thinks that Delta requires that Echo denies that Foxtrot affirms that the NATO alphabet is a waste of space.
With the possibility of this sort of nesting being part of the interest – although I did not come across it in reference 2.
I dare say people have looked into how much nesting one can cope with. For myself, I can just about manage the first example above, but completely fail on the next two. Maybe they burst some limit on working memory, or something of that sort. So what does one do with a sentence that makes sense in logic but which cannot be grasped by the living brain? If I read a long paragraph, perhaps made up of long sentences, that might be a bit tricky, and certainly exceeds working memory, but it might well make sense in the way that the sentences above do not. So what is the difference?
Another complication is that if someone reads one a long sentence, one might well be able to play it back, without having understood it. I associate to the way that people who are thinking about something else, can play back to you what you have just said to them, without having taken it onboard at all – which must be some kind of parallel processing. But I digress.
As an example of failure of comprehension, I offer the Wagner example which occupies a good chunk of the middle of reference 2, starting at page 107, that is to say my belief that Wagner died happy. Aside: I wonder today whether Davidson was a Wagner buff, or did he borrow the example from a colleague?
I have added the emphasis in the quote which follows, in an attempt to make things clearer.
‘… We now need to distinguish two related but different asymmetries. On the one hand, there is the familiar difference between self- and other-attributions of the same.attitude to the same person: my claim that I believe Wagner died happy and your claim that I believe Wagner died happy. If these claims are put into words, we have the difficulty of deciding what pairs of utterances are suitably related in order to guarantee that the claims have the 'same content.' On the other hand, we may consider my utterance of the sentence 'I believe Wagner died happy,' and then contrast my warrant for thinking I have said something true, and your warrant for thinking I have said something true…’.
But, unfortunately, it didn’t make it clearer. I did not, for example, see what utterances he was talking about. And furthermore, why did bringing utterances into the first asymmetry not reduce it to the second?
I tried again on the very last paragraph, on page 111, snapped in entirety above. I chopped it up into pieces and grappled with each piece in turn. Some of the pieces I understood and was content with, other pieces I was not content with, and others I did not understand. And I completely failed to get the big picture.
Aside: for a change, I had made use here of the Word comment feature, which was sophisticated enough, but perhaps a bit much for my modest needs. It will be interesting to see if I ever come to make a habit of using it.
Other matters
I complained above of Davidson’s use of philosophical jargon. From which I associated to the jargon ascribed to the police of Los Angeles by our holiday watching of Bosch and Ballard from Amazon Prime. It seems that said police litter their conversations with the numbers associated with certain events or procedures, presumably the numbers of sections or paragraphs in the police code, in whatever form that might take. They also litter their conversations with a large number of expletives, which I find irritating. Odd people these Americans: expletives and gory medical images in, female breasts out.
At another point, I associated to a lesson that I learned in my days in population estimates: it is easy to ask a question like ‘how many people live in Britain’, but getting to the right answer can be unexpectedly complicated – even without getting into the statistical vagaries of the numbers involved. Just for a start, what exactly do you mean by ‘Britain’? What about the Channel Islands? And then, a bit later on, you might get into all the US servicemen, quietly living in their bases, scattered across the country. It goes on.
I went on to try Gemini on the matter. In the first instance, he just answered the question, without embellishment. But he was able to fill in the detail well enough, when prompted.
He ended by turning up reference 15, the sort of detailed guide which was not published in my time with the relevant bit of what was then OPCS, now ONS.
He then rather caught me out with:
I'm curious – given your time in the unit, did you find that the biggest challenges were the technical, procedural shifts in methodology, or was it the constant pressure to make those figures meaningful for actual policy decisions?
Which I was hard put to answer. Gemini is certainly getting better at conversation. And he was not to know that I was interested in all the small print, until I starting digging a bit.
Along the way, I was reminded that the 1939 National Register was created ten years to the day before I was born. And lived on into my time as the National Health Service Central Register, then housed in the splendidly named Smedley Hydro in Southport, once a rather grand watering hole. For the register then, see reference 14. For the register now, see the very end of reference 16.
I mentioned above the circular possibilities of propositional attitudes. Which prompted me to try Google Images on the snap above taken from Gemini’s take on a Sanskrit grammar. He got top marks with: ‘The image displays the first 10 aphorisms (sūtras) of Pāṇini's Aṣṭādhyāyī, specifically from Chapter 1, Quarter 1 (Adhyāya 1, Pāda 1)’. He goes on to supply the text version of the image.
Conclusions
Attempting to read the paper at reference 2 has been only slightly more successful than the previous attempt at reference 1. On the other hand, it has provoked interesting speculations about the nature of conscious thought and about how better to tackle it.
But is that enough? I failed to understand much of what I was reading and I failed to get to the end of either paper – to the point where I wonder whether this particular game is worth the candle. I also suspect that things have not moved on much in the forty years since these papers were written; that papers written today by Davidson’s successors would be just as impenetrable, just as inaccessible to the lay reader.
So once again, I have failed to properly connect with the work of a professional philosopher. The head of a philosophy department at a university would still get a hard time if I was sitting on the funding committee.
But then again, philosophy departments do not burn up money in the way of science departments, they are not greedy, going no further than a modest diet of very important conferences in pleasant foreign parts. So maybe I would let them be!
References
Primary references
Reference 1: Knowing one's own mind – Donald Davidson – 1987.
Reference 2: First person authority – Donald Davidson – 1984.
Reference 3: https://psmv5.blogspot.com/2023/09/swampman.html.
Secondary references
Reference 4: https://en.wikipedia.org/wiki/How_many_angels_can_dance_on_the_head_of_a_pin%3F. Angels on pin heads.
Reference 5: https://psmv6.blogspot.com/2026/06/reading-brain.html.
Reference 6: https://en.wikipedia.org/wiki/Epistemology. A regular menagerie of different schools of thought. Davidson talks a lot about the sceptics mentioned here.
Reference 7: https://psmv6.blogspot.com/2026/07/jasmines.html.
Reference 8: https://en.wikipedia.org/wiki/Propositional_attitude.
Reference 9: https://iep.utm.edu/prop-ati/. For more of propositional attitudes.
Reference 10: https://plato.stanford.edu/entries/prop-attitude-reports/. Plenty of stuff about them in PLATO, but I did not find an entry called ‘propositional attitude’.
Reference 11: https://en.wikipedia.org/wiki/Donald_Davidson_(philosopher).
Reference 12: https://en.wikipedia.org/wiki/A%E1%B9%A3%E1%B9%AD%C4%81dhy%C4%81y%C4%AB. The Aṣṭādhyāyī. Found by Google, rather more helpful than Bing on this occasion.
Reference 13: Principia Mathematica – Whitehead, Russell – 1910-1913.
Reference 14: https://en.wikipedia.org/wiki/National_Registration_Act_1939.
Reference 15: Population estimates for the UK, mid-2021: methods guide – Office for National Statistics – 2022.
Reference 16: https://psmv6.blogspot.com/2026/07/fake-202.html.
Reference 17: https://en.wikipedia.org/wiki/Skepticism.
Reference 18: https://psmv3.blogspot.com/2016/08/descriptive-experience-sampled.html.
Reference 19: https://psmv2.blogspot.com/2015/08/aryans.html.
No comments:
Post a Comment