This site sets cookies to help me tweak settings to make things easier for visitors. If you object to cookies, adjust your browser settings to reject them.
Saturday 6th February 2010
Chang combines an historical account of the construction of temperature scales with an analysis of the implications for the Philosophy of Science.
I found the history fascinating, and the analysis acute.
Constructing a temperature scale seems to involve circularity, because it appears we can't construct one unless we already have one.
We need to choose some physical quantity to measure temperature. That has usually been the volume of a fluid, though length of a silver rod, and pressure of a quantity of gas have both been used. However different physical quantities give different scales which agree only at the fixed points used for calibration.
Fixed points too presented serious difficulties. We need processes that always occur at the same temperature. How are they to be recognised unless there is already a temperature scale? Melting point and boiling point of water were favourites, but water neither melts at a fixed point, nor boils at a fixed point; the boiling is especially erratic. One of the hazards of the Chemistry laboratory is 'boiling with bumping' when a liquid is heated to above what we like to call its normal boiling point, and then a great deal evaporates quickly in a minor explosion. Eventually physicists settled on the temperature of steam over boiling water at a pressure of 76 cm of mercury, but did so only after more than a century of debate.
Temperatures outside the range of a mercury thermometer presented immense difficulties. For a while many people denied that it was possible for mercury to freeze, using the Aristotelean argument that it is 'essentially' a liquid. When therefore it did freeze in the thermometers used in an expedition in Siberia, ridiculously low temeratures were recorded because people mistook the contraction that takes place when mercury freezes for a large contraction of what they assumed to be still liquid mercury.
When Wedgewood wanted to measure temperatures in his furnaces, temperatures that would melt most metals, he had to invent a special scale of his own based on the contraction of clay when heated. It took many years to link Wedgewood's scale to the Centigrade or Fahrenheit scales.
A fascinating sub plot is the experiment to show that heat is a radiation by using special mirrors to bring it to a focus. The odd thing about that experiment is that it can also be used to bring 'cold' to a focus. That observation was so unwelcome that the effect is usually ignored, though it was shown to be genuine by being repeated in the 1980's. (I think I ought to stick to one tense in that sentence, but can't decide which it should be).
Chang concluded by proposing a study that he calls 'Complementary Science', concerned with the strange byways that conventional science has to neglect in order to concentrate on the research that appears most fruitful at the moment.
Thoreau, once described as a 'Yankee Diogenes', was an extreme individualist who found mid-nineteenth century life over complicated, and thought people spent too long acquiring property that they could easily do without. He preferred to live a simple life which he could support by working only a few weeks each year, so that he had plenty of time to read, write and think. From 1845 to 1847 Thoreau lived in a hut in the woods on the shores of Walden Pond, near Concord, in Massachusetts.
He built the hut himself, on land owned by his friend Emmerson, at a total cost of just over 28 dollars. Walden is his account of his life there, interlaced with a variety of thoughts that occurred to him during that time.
He supported himself partly by doing an occasional day's paid work, sometimes as a surveyor, and sometimes just as a labourer, and partly by growing assorted vegetables, mainly beans and potatoes. Some he ate, but most of the crop he sold. The prices intrigued me, and so did the measures he used. He recorded selling potatoes at 50 cents per bushel for the large ones, and half that price for the small ones. He kept a firkin for his own use.
That inspired me to check what a those measures are. A bushel appears to be eight gallons, and eleven bushels make a firkin. Thoreau's favourite measure of distance was the rod, which is 5.5 yards = a quarter of a chain.
When I read Thoreau I usually had beside me my old 1977 Science Diary that gives conversion factors for most of the imperial measures
At first I was irritated by the stridency of some of Thoreau's strictures on what he considered the over-sophistication of civilised life, but that irritation was diminished when I realised that he was just explaining his personal preferences. He did not want everyone to follow his example, indeed he thought the world would be a dull place if all behaved in the same way, and urged his readers not to be bound by custom and to work out each his own way of life. I say 'diminished' because I still think Thoreau underestimated the almost universal prefence for the complications of civilised life over the simplicity of the primitive. Civilisation evolved in response to the discontent of the uncivilised with the primitive simplicty of their lives.
He seemed to regard life in the hut as a proof of principle rather than a settled way of life, because after two years he left it, considering that he had proved his point.
I had to force myself to read Walden all through without skipping. It is a strange mixture of acute observation, epigrams that might have been written to go into an anthology of telling quotations, and tortuously meandering metaphysical wool gathering.
Thoreau was a close observer of nature. Studying the shapes made in the sand by rivulettes of rain water trickling down the railway embankment, he described the fascinating complexity of what we should today call a fractal structure. Unfortunately the records of his observations ammounted to little more than a clutter of unordered data, because he had no coherent theory that he could use to put them in order. That was not because he had no theory, but because he was in the thrall of a woolly Emmersonian proto-theory, that all Nature is One Wise and Wonderful.
Civil Disoberdience, much shorter than Walden arose from Thoreau's brief imprisonment for refusing to pay taxes as a protest against the institution of slavery, and against the American war against Mexico.Civil Disobedience has been quite influential, apparently giving Gandhi the idea of passive resistance.
The last 150 or so pages of the book are taken up with essays by various commentator about Thoreau and his works. Some are admirably clear, including, surprisingly, Emerson's, but others I found very tedious, so that it was a struggle to finish.
At intervals during the year, I tried to learn more about Javascipt, and later Frames Forms and cgi programming with the aid of various books borrowed from the Library.
The books I've read have been quite unremarkable, so I don't give details.I mention this part of my reading only explain why I've read rather less other material than in previous years.
Lipton argues that much, if not all, scientific reasoning is what the title says. He was attempting to rehabilitate inductive inference
I bought the book after reading obituaries of Lipton, saying among other things that he was an outstandingly good lecturer, whose student audience once showed their appreciation by showering him with rose petals at the end of a course of Lectures. He was said to be particularly skilled at turning the incoherent half developed arguments of his students into lucid prose that the individuals in quesion acknowledged as expressing their thoughts perfectly. They invented the verb 'to Lipton' to refer to the procedure.
He died suddenly in 2007, aged 53, having for some years been professor of the History and Philosophy of science at Cambridge and a fellow of King's.
The book is proved less of a revelation than I had hoped. The pedestrian prose has none of the sparkle the obituaries led me to expect. In some ways its approach, though not its conclusions, reminds me of some of the work published in the 1920's and 1930's. Lipton seems to have thought that one should be able to find formal rules for identifying an explanation. That is not possible, because no such rules can be established antecedently to scientific enquiry, since the questions of what methods of investigatiion are fruitful, and what sorts of clues are worth following up, are themselves things we can only learn by experience.
Lipton might have understood that better had he paid more attention to C.S. Peirce, who was mentioned only once in a bare reference that was not followed by any discussion of the material refered to.
Lipton thought that an explanation should be in terms of cause and effect. Cause atracted him because he thought explanation is typically partial. We don't usually just explain why something happened, we explain why one thing happened rather than another. In other words, explanations are usually contrastive.
Cause entered the philosophical vocabulary through Aristotle's notion of arkai. Aristotle defined cause as an explanation of change, distinguishing four sorts of explanation, material, formal efficient and final causes. (see Chapter 2 of my Philosophy notes) The modern notion of cause is Aristotle's efficient cause. If we follow Aristotle in defing cause as a type of explanation, it woiuld be circular to define explanation as the production of causes.
Lipton conceded that the notion of cause is itself problematic, though he didn't seem to apreciate how problematic. I suspect he may never have read Bertrand Russell's 1912 paper On the Notion of Cause in which Russell argued cogently that 'cause' has no place in science - or at least no place in Physics. Lipton thought the difficulties we have with 'cause' do not prevent our using it to analyse 'explanation'. What he overlooked was the nature of the difficulties with 'cause'. The central difficlty is distinguishing a relation of cause and effect from an accidental relation of regular succession. What makes a regular succession into a relation of cause and effect is the ability to provide an explanation of the effect. Thus cause has to be elucidated in terms of explanation, making the use of cause to analyse explanation viciously circular.
Lipton's examples of explanations were relatively few, and usually involved explanations of particular events, such as someone deciding to study at one University rather than another. They had no clear relevance to scientific thought.
I think it is a mistake to look for one logical structure that must be present in every explanation. To explain some event is to incorporate it into some sort of story that makes it appear to fit harmoniously into our system of knowledge. That usually makes what we are explaining seem less surprising than it otherwise would.
Although that is very vague, I think it is about all we can usefully say about explanation in general, because there are many ways of explaining, so there is no simple and precisely definable pattern covering them all.
In places Lipton seems to confuse the problem of testing a theory and justifying belief in it, with the very different problem of developing the theory in the first place. His work lacks the subtlety and analytical acuteness I expect from someone who has spent much of his life among academic philosophers. The History and Philosophy of Science can easily become a cul de sac where people plod on largely unaffected by the debates that engage the wider Philosophical community.
I attended the course of lectures out of which this strange book developed, so it was particularly interesting to me. I bought it when it was published in 1975, but on opening it again realised that I'd previously read barely half its 157 pages. Much of it was not easy reading, with frequent references to numerous numbered sentences, but this time I managed to read it all.
For Lewy sentences all had, or at least could be made to have, precise meanings, and philosophical conclusions could be established by deploying quasi mathematical proofs.
Lewy was concerned to refute the claim, quite common in the middle of the last century, that logically necessary propositions are linguistic, so that, for instance 'A vixen is a female fox' provides information about the use in English of the word 'vixen' and the phrase 'female fox'.
That claim Lewy refuted with impressive overkill. He the turned to discuss modal logic, arguing for a medieval distinction between modalities de re, and de dicto. He applied that to the so called 'Paradox of Analysis' that every identity statement is either false or trivial. That is generated by considering propositions of the form 'a = b' It has often been asserted that, if that is true, it is equivalent to 'a = a'.
Finaly he discussed entailment and various attemopts to avoid the conclusions that a contradiction entails anything at all, and that a logical truth is entailed by anything at all.
Reading all that inspired me to add five pages to chapter 5 of my Philosophy notes.
I saw a reference to this book while making an Internet search for information about the architect Bertholdt Lubetkin. Louise Kehoe was the youngest of his three children. She describes a turbulent childhood, followed by a search for her father's antecedents.
Berthold Lubetkin was born in Russia and after various adventures and misadventures came to Britain around 1930 setting up a successful architectural practice. Towards the end of the 1930's he suddenly gave up architecture and took up farming, buying an isolated farm in Gloucestershire.
Throughout his life he remained a staunch Communist and a loyal defender of the Soviet Union, and expected his children to defend the same point of view.
Lubetkin always told his children that 'Lubetkin' was not his original name, saying he had changed it to make it easier to enter Warsaw Univerity after he fled from Russia following the Bolshevik revolution, in the course of which all his family had perished, or so he said.
Kehoe discovered that that was not true. 'Lubetkin' really was his name, and he seems to have claimed otherwise to conceal the fact that his family was Jewish. His parents lived in Poland till 1941, when they perished at the hands of the Nazis. One cousin was still alive in the 1990's, living in New York where Kehoe visited her to learn that Lubetkin himslf had visited her there a few years before his death in 1990. He had told the cousin that he had no children, and had told his children nothing of the cousin.
Lawrence biefly explains a classification of human personality, provides some statistics on the distribution of various types in the population of the United Sates, and then suggests implications for educational practice.
The classification, developed by Isabel Myers-Briggs is an extension of Jung's, which suggests one should treat it cautiously.
Although mainly written by Lawrence, the book also contains material by others, including reprints of an essay and an introductory pamphlet, both by Myers Briggs, and an extract from the Manual of the Myers Briggs Type Indicator.
Personalities are assessed on four dimensions.
First is Jung's widely known, but apparently almost as widely misunderstood, distinction between extravert (E type) and introvert (I type). An extravert is primarily interested in the outside world, and introverts are primarily interested in the inner world of their thoughts and ideas; in neither case does 'primarily' mean 'exclusively'. Note that this does not imply than an extravert need be a noisy person trying to be the centre of attention, or that an introvert need cower in the corner of the room on any social occasion. An extravert could well seek to re-organise the world with quiet determination, and an introvert might loudly defend the integrity of his internal space. The crux of the distinction is the direction of their primary interests, outward or inward, not the style with which they pursue those interests.
The way we handle perception defines another dimension. Perception involves a combination of experience and interpretation. Those referred to as 'Sensing' (S) types take what they sense as primary, and treat reflection on it as a subsiiary activity. On the other hand for Intuitive (I) types thoughts and theories are primary, and the function of perception is to aid the development of thought.
Resolution of problems usually involves both thought and feeling. For T types thought and logical analysis are the primary tools, while F types place more reliance upon feeling.
The fourth distinction differentiates people according to their attitudes to the ordering of their lives. J types prefer to have things orderly and cut and dried, while P types like to leave things open ended and to delay decisions in case more information becomes available. A J type friend of mine has seven different breakfast menus, one for each day of the week, and adheres to the pattern rigidly. I have the same things for breakfast every day. Both of us get up at the same time every morning, and set the table for breakfast the night before. Anyone who is surprised to read that is an S type.
My personality type seems to be: INTJ, apparently quite a rare combination found in only about 2% in the American population, although quite common among Mathematicians and Scientists.
What does one make of such a theory? Is is really a theory at all ? One could treat a classification as a sort of conceptual filing cabinet. Is a clssification falsifiable? I can thnk ofone reason one might reject a classification, if there were cases to which it could not be applied. However the possibility of classifying everyone does not imply that that classification is particularly interesting, and there could be uses for a classification that could not be applied to every case, provided it applied to many.
There are several ways in which a classification could be interesting. It might act just as a mnemonic, helping us sort out our observations to our saitisfaction, and possibly also helping us formulate speculations worth investigating. However that may not amount to very much. I once read of experimnents in which groups of people were set to solve a problem, and some groups were given a keyword while others were not. The groups given a keyword made more progress than the others, even though the keywords were chosen at random from a dictionary without regard to any reference to the subject under discussion.
A more significant way a quality or classification can be interesting is by being projectable in the sense of suporting inductive inference. Popperists say there is no inductive inference, because they don't know what induction is supposed to be and enjoy the denunciation of intellectual sin too much to try to find out. (see chapter 6 of my Philosophy notes, available from the Philosophy page of this site).
Once the classification has been explained, examples are provided in which someone's psychological type is offered as an explanation of their behaviour, in particular of the difficluties some children have at school. That makes me uneasy for three reasons. To some extent the personality type assigned to someone is a description of that person's behaviour, so to use type to explain behaviour is akin to explaining the behaviour in terms of itself. There is also an implicit assumption that someone of a certain type will be an extreme caricature of that type - that a T person tends to be unaware of feelings, and and F person does not think. Finally type is assumed to be fixed. It may be, but the assumption that it is needs to be supported by evidence. I assume the phrase Tiger Stripes in the title refers to the saying that a tiger does not change its stripes, seeming to beg the question whether or not we can change our personality type.
Behind the Myers-Briggs classification there are several assumptions that are testable, or at least ought to be. The one that appears to me the most dubious is that everyone will, on each dimension, incline more to one side than to the other. The possibility of an even balance seems to be ruled out. Tests I've found on the Internet score from 0 to 5 on each dimension, so there is no score corresponding to a mid point. It is also assumed that classifications are stable, so that the same person will be given the same classification when tested at different times. I gather there are supposed to be some data supporting the latter assumption, but I haven't come upon any reference to any support for the former.
I borrowed the book from the Library to help me prepare for a discussion about Darwin at a meeting of the U3A Science and Technology Group.
Although it contained some biographical material, and summarised Darwin's own work, the book centred on Darwin's influence. Illustrations were few and in my opinion ill chosen. Most showed people or buildings, and I don't recall seeing a single illustration of a plant, animal or fossil. As Darwin's argument often hinged on details of the anatomy of creatures likely to be unfamiliar to most readers that was unfortunate.
I hadn't previously realised that support for natural selection as the principal engine of evolution did not become the consensus until well into the twentieth century, with many biologists inclining to the inheritance of acquired characteristics. Indeed Darwin himself seems not to have ruled that out as one factor.
The weakness in Darwin's position was that in his day there was no well established theory of heredity. Darwin himself thought that each of the organs in the body might release tiny buds, and that budy from the various organs might somehow be gathered into the germ cells, but that was only conjecture on his part.
Darwin was also hampered by the poverty of contemporary statistics, which prevented him analysing some of his own results in plant breeding. He sought help from his cousin Francis Galton, who could not solve Darwin's problms, but was inspired to undertake research of his own in statistics. Galton helped to develop the theory of correlation
Although it is quite a short volume - comprising only 158 pages, I found this book very useful because it was lavishly illustrated - a welcome change from Peter Bowler's work.
Baggini has collected 100 short storie - he calls them 'thought experiments' - that have been used to illustrate philosophical problems. The stories themselves usually take up less than a page, and each is followed by a couple of pages by Baggini indicating what conclusions we might draw from it. I should have preferred more stories and less from Baggini, so the stories could speak for themselves.
Several of my favourites were missing, notably the stories John Wisdom used to tell, such as William James and the squirrel, Wittgenstein and the white rabbit, the ewe lamb, watching one's own funeral, and the gardener.
Fortunately I didn't have to pay for the book, at least not directly. It came without extra charge as an incentive to me renew my subscription to The Philosophers' Magazine, which Baggini edits.
This is yet another book that I've had for more than 30 years, and have never before read in its entirety
Over the decades I'd forgotten a great deal of my abstract Algebra, partly because I originally tried to read too fast. This time I'm content to prgress by as little as one or two pages per day in when dealing with the tickier parts.
Progressing at two pages per day is not the same as reading only a couple of pages per day. I read through much of the material several times. At the first attempt I'll get an idea of what is to come by skimming quickly through a few pages ahead of what I've previously read more carefully. At the next stage, usually the following day I re-read that material carefully amd try to think of some examples. Finally at the third attenpt I try to understand the details of the proofs. So on any particular day I'll usually start in the morning wth the third reading of a couple of pages, then, either immediately afterwards, or possibly later in the day I do the second reading of the next page or two, followed by a skim through the following few pages.
My strategy seems to have worked. I now have a reasonable idea why there is no general formula to solve quintics and higher polynomials in terms of radicals. Before I simply remembered that I'd plodded through proofs and followed then at the time, without remebering anything of the details. I've also discovered the algebraic significance of quaternians.To oversimplify, there are only three algebraic systems that contain the real numbers, that are such that every element in the system is a root of a polynomial with real coefficients, and which permit division by any non-zero elemnet. The three algebras are: the reals, the complex numbers and the quaternians.