William H. Newton-Smith (1943–2023)

William (Bill) Newton-Smith was a renowned Canadian philosopher of science who spent his career largely in Oxford and then at the Central European University in Hungary.

Newton-Smith was born in Orillia, Ontario and completed a B.A. in mathematics and philosophy at Queen’s University, Canada. In his second year as an undergraduate, he decided he was a logical positivist. In the interview for an exchange scholarship with St Andrews, in which he would be successful and at which he would spend his third year, the head of the Queen’s philosophy Department remarked that he had thought logical positivism was dead. With the confidence of youth Newton-Smith replied that the fact that it may not be popular was no reason to think it wasn't true!

During his year at St Andrews he fell under the spell of the later Wittgenstein. Upon graduating from Queen’s in 1966, he went to Cornell to do a PhD, as Norman Malcolm had photocopies of all of Wittgenstein's unpublished works. Unimpressed with Malcolm, Newton-Smith transferred to the analytic philosopher Max Black. He thought Black a great supervisor, lively and clear—just the sort of teacher and mentor he would himself become. But Newton-Smith had already decided to transfer to Oxford. Black didn’t want him leaving Ithaca without a Cornell degree, so he secured funding for him to do an MA over the summer.

Once at Oxford, Newton-Smith became a leading figure there. He finished his DPhil under the supervision of the logician Arthur Prior. In 1970 he was elected a Fellow of Balliol College. He was a pillar of the College and the University, taking on important administrative positions for both. He was a beloved undergraduate tutor, lecturer, and supervisor of an astounding number of DPhil theses. He could make logic fun and had a fine nose for what was a good (and bad!) argument in a student’s work.

Newton-Smith made important contributions in several areas of philosophy, especially the nature of time and the general philosophy of science. Perhaps his two most important works are The Structure of Time (1980) and The Rationality of Science (1981). The latter appeared in the middle of the realism debate, a hot topic at any time but particularly so in the late 70s and early 80s. Even after four decades it stands up well and gives an excellent snapshot of his realistic approach to philosophy.

Newton-Smith was a scientific realist; he had no qualms about the theoretical entities postulated by science, though he did have an interestingly radical solution to one problem that presses in on the realist. He was also a rationalist; he believed that science has clearly progressed and part of the philosopher’s job is to explain or make sense of that progress. As Newton-Smith saw it, there are a number of challenges which face the traditional rationalist, including: overcoming the problem of incommensurability (the same term could have different meanings in different theories); it must be shown that science does indeed have a goal; and explaining how the normative methods of science lead to that goal. Finally, it must be shown there is a good fit between the prescriptive philosophy of science and the descriptive history of science, for only then can we give a satisfactory account of real scientific change.

To meet these challenges, Newton-Smith proposed new accounts of facts, reference, verisimilitude, and normative methodology which he hoped would overcome any difficulties standing in the way of adopting a realist and rationalist account of science. He described his position as ‘temperate rationalism’. There seem to be a couple of reasons for the tempering label. First, Newton-Smith backed away from the belief that every proposition is either true or false; he did this to save realism from the problem of underdetermination (there are always multiple theories that can account for the available evidence). We should note that calling this move ‘temperate’ seems an understatement; it is actually quite radical. Second, his normative rules for doing science are relatively loose in their characterisation, and these norms are allowed to evolve over time. Such liberalisations are a move toward moderation.

However, the idea of being moderate, temperate, and so on should be viewed with a sceptical eye. One of us (JRB) recalls teasing Newton-Smith about this, citing Bernard Shaw’s remark about the moderation of the English middle class: they are moderately honest and moderately faithful to their spouses. In Newton-Smith’s case, however, there is little that is moderate or temperate about his realism. With one big proviso—the denial of bivalence—he was pretty much a classical realist who set out to justify realism and the rationality of science and was quarrelling with other realists over the important details.

Let us turn now to some of these details, beginning with his account of scientific realism. Newton-Smith began by showing that the theory/observation distinction does not hold up, a move which then allowed him to attack the instrumentalist. The aim of theories, says the instrumentalist, is merely to save the phenomena (i.e. get the observable predictions right). The notions of truth and reference do not apply to theoretical sentences—such sentences are instruments that are better or worse, not propositions that are true or false. But with the failure of the theory/observation distinction, the notions of truth and reference would not apply to the so-called observation sentences either. And with this absurdity out goes instrumentalism. Relativism, too, is rejected. The doctrine that proposition could be true for one culture/theory and yet false for another cannot be coherently spelled out. Culture C₁ (or theory T₁) could say ‘P is true’, while C₂ could say ‘P is false’. We might think this is short for ‘P is true in some cultures and P is false in others’. Now imagine C₃ that says ‘P is false in every culture whether they believe it or not’. Then we would have: ‘P is true in some cultures and false in every culture’. These sorts of oddities can be generated endlessly. That sounds like a threat, so we’ll stop.

Included in Newton-Smith’s account of scientific realism are several ingredients, but two are central and standard. One is the ontological claim that our theories are true or false in virtue of the way the world is independently of us. A second is the epistemological claim that we can have good reasons for choosing one theory rather than another as being more likely to be true (or closer to the truth). In addition, Newton-Smith included two more ingredients: the causal ingredient says that evidence for a theory is evidence that the entities that it postulates really exist, and the thesis of verisimilitude which says that the historical sequence of theories is getting closer to the truth. These were both major innovations in the characterisation of realism and, needleless to say, controversial.

Perhaps the most important consideration which holds back general acceptance of scientific realism is the possible underdetermination of theories by data. Suppose that theories T1 and T2 make exactly the same empirical predictions, including all future predictions, and yet contradict one another over some theoretical claim. Suppose further that as well as getting all of their observation claims right, they also do equal justice to all non-observational considerations; that is, they are equally ‘simple’, they ‘harmonise’ equally with non-scientific theories, and so on. It would seem that there can be no rational choice between them. This presents an obvious problem for scientific realism. The epistemological aspect is undermined since here we have a pair of rival, contradictory theories, T1 and T2 and we cannot possibly choose rationally between them. The aim of science should not be the truth, since it is unattainable.

The responses to the underdetermination problem have ranged from embracing it and abandoning realism to claiming that the contrary nature of the two theories is only apparent (that they must be equivocating). Newton-Smith's solution to the problem is disarmingly simple and extraordinarily radical. He denies that there is any fact of the matter that the two theories disagree over. In doing so, he denies that there can be such a thing as an inaccessible fact (i.e. one which cannot be ascertained in principle), and in consequence he denies the universal validity of the law of excluded middle. Calling it the arrogance response, he

writes:

… to posit a fact making p true or false is to admit the existence of an inaccessible,

gratuitous fact the presence or absence of which could never be discovered. Why, we

might ask, should we bother to assume that there is a matter of fact at stake at all? Why

assume that the world is determinate with respect to p? To move in this direction is to make what I call the arrogance response to underdetermination. On this response we drop the assumption that there is a fact of the matter at stake with regard to any undecidable

sentence.

This smacks of verificationism, which Newton-Smith would not have liked. It seems as if a proposition has a truth value if and only if it can in principle be checked. Logical positivists were his early philosophical heroes. To be a realist is usually understood to be a move away from that sort of view. Newton-Smith was determined to have it both ways. Needless to say, it was a controversial view. But it won converts who felt the challenges to realism were so difficult that something new, such as Newton-Smith’s proposal, was needed. Looking back 40 years, it is safe to say his proposal did not prevail, but it was a wonderful stab at an enduring mystery.

Newton-Smith was a champion in the fight against authoritarian regimes. His work in politics and education was just as important as his work in philosophy. In the 1970s the beautiful medieval city of Dubrovnik became the home of an important effort to mitigate the effects of cold war on intellectual life. Scholars from east and west happily mingled in neutral Yugoslavia. The annual philosophy of science conference was the first of these, started by the physicist and philosopher Ivan Supek who convinced Werner Heisenberg to participate. It has been going strong since then, even during the violent breakup of Yugoslavia when the city was under siege. Covid proved to be stronger than sniper fire, but after a short hiatus, the conference is back.

Newton-Smith quickly became the main organiser, inviting people from all over. Later Kathy Wilkes, his Oxford friend and colleague, took on much of the organising work. This annual conference has long been famed for its wonderful friendly atmosphere. It was no doubt deliberately cultivated during the cold war, but has managed to become the well-entrenched culture. Newton-Smith had much to do with the amiable and easy going nature of things—long coffee breaks, longer lunches, and even longer dinners with non-stop discussions creating something that paradoxically stimulated and relaxed simultaneously.

Scholars debated the issues of the day and often issues that were unknown to those on the other side of the east–west divide. This is where the two authors of this obituary, along with hundreds of other philosophers, learned about Polish positivists, Romanian methodologists, Bulgarian realists, and novel Russian takes on Kuhn. It could be perplexing, frustrating, and enlightening. Casual conversations were amusing: Some from the west would denounce Ronald Reagan while some from the east would profess a fondness for Margaret Thatcher. All this was possible because of the warm atmosphere created and maintained by Newton-Smith.

Newton-Smith was also co-founder of the Jan Huus Foundation, which did important work, with Václav Havel and others in Czechoslovakia, for freedom of expression and inquiry. The Foundation made available the means to create samizdat publications, by smuggling in both mimeograph machines and academics with their ideas. He was arrested in Prague as he gave a lecture to a gathering of philosophers in a private flat, interrogated by the secret police, then expelled, taken to a West German border crossing in the middle of the night. One of us (CM) was sent on an ‘cloak and dagger’ mission by Newton-Smith in the mid-80s and she has been dining out on it since.

In 2005, Newton-Smith took early retirement from Balliol and Oxford, in order to help found the Central European University and continue running the educational mission of the Soros Foundation. He then spent most of his time founding projects for the Open Society in the republics of the former Soviet Union. In Kyrgyzstan, he played a major role in the creation of the American University of Central Asia and he was involved in the founding or rebuilding of over 35 universities on behalf of the Soros Foundation. This caused him to travel widely and on one of these missions, he fell off a camel and broke his foot, the source of merriment amongst his friends.

After years of vital work in education in the former eastern bloc countries, Newton-Smith and his partner Nancy Durham moved to their beloved country home near Builth Wells, Wales (a village known to generations of Oxford logic students as it figured heavily in Newton-Smith’s examples in tutorials and in his logic textbook). There they grew the most wonderful lavender and transformed the farm into a commercial enterprise, making fragrant creams and soaps. He was hugely proud of his two girls, Rain and Apple Newton-Smith and gloried in his grandchildren.

The journal International Studies in the Philosophy of Science was created by Bill Newton-Smith and Kathy Wilkes in the mid-1980s to serve the Dubrovnik conference and the wider community. It has been flourishing ever since.

Bill Newton-Smith died on 8 April 2023. His generosity, wicked sense of humour, and tremendous contributions to philosophy and to the freedom of education and inquiry are a cause for celebration. He would have liked us to raise a glass of something good to him.