Conventionalism is a philosophical concept whereby some principles or propositions, both cognitive and ethical-political are based on conventions , agreements or choice (even implicitly) that can not be assessed in terms of truth or falsity. This conception has been the subject of deep analyses since ancient times, extending from periods of greater or lesser originality to contemporary scientific and philosophical studies.
Here we are faced with a form of methodological conventionalism: the scientist is guided by the idea of simplicity. Simplicity is synonymous with truth. This way of proceeding is based both on habit (historically it has always been proceeded in this way), and because, on a strictly pragmatic level, this approach is the only one that enables science to proceed.
Modern conventionalism can be described as a school of thought that emphasizes the conventional nature of some fundamental concepts of mathematics and physics. It starts from the discovery of non-Euclidean geometries and relativization of space and time, determined by the physics studies in the early XX century. With the non-Euclidean geometry it was called into question the spatial intuition, which stems from common sense and the psychological and physiological investigations, understood as the foundation of geometry.
The reflections of the scientist and philosopher Mach (in the works "The Analysis of Sensations" and "Knowledge and Error"), the philosopher and historian of science Duhem and especially the great mathematician J.-H. Poincaré contributed deeply to the conventionalist analysis of the development of sciences. Mach, for example, with his radical empiricism, reduced concepts and scientific theories to pragmatic means able to organize sense data in the most economic and predictively effective way. Duhem developed Mach epistemology characterizing physical theory as a system of propositions derived from certain principles that represent experimental truth in a simple and rigorous manner.
Poincaré gave an important contribution for reflection on conventionalism either denying the validity of the Kantian theory, which considers the Euclidean geometry an ‘a priori’ science, or rejecting the idea that non-Euclidean geometry (and geometric systems in general) be empirically verifiable.
These are simply conventional truths, to which experience can not impose any falsification. Take, for example, the acceleration of a body that does not depend on the position of this body, of the neighboring bodies and their speeds. This proposition, to be verified completely, would require that, after a certain period of time, all bodies in the universe would return in the initial positions with their initial velocities. Only in this case it could be verified if they rejoined the initial trajectories. In fact, the generalized principle of inertia implies that, at equal positions and speed, the accelerations are the same. Such a principle can never be falsified empirically.
The discoveries of the late nineteenth century seem to condemn the principle of Carnot, of Galilean relativity, that of action and reaction, the principle of conservation of mass. Poincaré argues that although these principles, on a strictly logical level, can be put back into play, after such an ad hoc rescue they would be completely useless. A principle has to be fruitful for the development of physics; when it no longer is, it has to be abandoned.
Conventionalist thesis type are also present in the work of R. Carnap in a certain period and more generally in logical positivism.
Carnap - in response to the influential objections of Quine on the opportunity to provide a clear distinction between analytic and synthetic statements on which positivism based its epistemology - would extend the conventionalist principle to some semantic aspects of language, with the proposal to consider analytic truths as the "meaning postulates", that is, conventional truths no further justified except by virtue of a pragmatic choice.
Forms of conventionalism are also present in the post-Popperian and post-positivist philosophy of science.
The most radical exponent of the "new criticism of science" was the French mathematician Edouard Le Roy, who advanced a form of nominalistic conventionalism, following which science has only an instrumental value of rule of action; the same scientific facts are a pure creation of the scientists, only the "brute facts" would manifest the character of objectivity. Poincaré takes the field against this extreme form of conventionalism, criticizing and leaving the distinction between brute and scientific fact: every fact, he argues, is a "creation of our spirit." To tell if a thing is true or not, it must be aware of the conventions on the basis of which the application acquires a precise meaning. A scientific fact is simply the translation of a brute in a statement expressed in a language more contented. The scientist is a creator in respect of the rules of the language he uses, but to experience remains the task of responding once the question has been formulated.
Duhem will argue with Poincaré on this point, advancing a conventionalistic position more marked: the task of the scientist and his creativity are not limited to language processing, but enter also into the theoretical field.
Gianrocco -
Are you referring to something such as the way the model of the atom has developed thus far? It has been refined over time with something that might be said to be akin to an Ockham's Razor-like approach?
Jim
In this paper on Ockham's Razor,
https://www.researchgate.net/publication/220151263_Ockhams_razor_empirical_complexity_and_truth-finding_efficiency
the author, Kevin Kelly, states that (Section 5), "Efficiency Implies Ockham's Razor."
Does "conventionalism" come from efficiency?
Article Ockham’s razor, empirical complexity, and truth-finding efficiency
James,
your interesting questions had me make considerations in some way related to your thoughts. So it came to my mind that the simple ideas are the obvious result of feelings, the atomistic bases of knowledge, while the complex ones are the result of a reflection of the mind that, under the juxtaposition of simple ideas (for example, a color, a shape, a weight) search to identify a substance (for example a table). However, complex ideas, according to Locke, can not guarantee the formulation of certain judgments, but contain only elements assessed in terms of probability. And if you move to abstract or general ideas (what Plato had indicated as pre-existing models to the human mind), in this case for Locke we are before simple names, and abstract metaphors to which it does not correspond any objective reality : we need them, they are useful to our thinking, but in their essence they are only linguistic signs required for communication between people.
On the way open by Locke there were various thinkers, including David Hume, one of the most important exponents of philosophical relativism, and later on John Stuart Mill, for whom all knowledge is of empirical origin and the method on which it is founded is induction. In fact, according to Stuart Mill, the same deductive procedures start from general premises that can be generalizations more or less based on empirical observations. The same logical and mathematical truths are generalizations of some experiences of space and of relationships between objects.
The inductive generalization is in turn based on the idea of the uniformity of nature and regularity of natural phenomena: the transition from observations on a limited number of cases to statements on all of the possible data holds only on the idea that nature has laws . And here's the key point: nature is really subject to laws, or are we to think that it has? And, in the first case, how is it possible for man to know? In the second case, how can man make assumptions that do not correspond to reality? But these questions lead to other ones: What is knowledge, and is it possible? and reality does really exist?
Thank you for contacting me. Gianrocco
Gianrocco -
I noted this in your response:
"And here's the key point: nature is really subject to laws, or are we to think that it has?" The scientific method assumes that we can and should test to determine the truth of 'laws,' and if results are found to be inconsistent, then the models change, but I suppose one generally assumes that there are such laws to discover. Perhaps those laws are different elsewhere, but there has always been a quest for them, and consistent experimental results, and consistency with mathematical models seem to give encouragement. The assumption that there are laws at least gives humanity a starting point.
Your question then is: "What are the most significant scientific contributions of conventionalism?"
Perhaps if simplicity were not often a convention, some studies may go well, but I think that it is far more likely that discovering workable solutions would not only be found less efficiently, but perhaps missed altogether.
In statistics, regression is often a useful tool. It is often known that important independent variables may go unknown, or at least unmeasured/unobserved, which may result in an intercept term ineffectively trying to absorb/adjust for this. But overparameterizing - using terms that might be specious - is a bigger problem. If future data/information/subject matter theory suggests the regression model needs to be expanded, that should be tested. This is a convention not always followed, but those who are paying attention often note the consequences.
In Gravity, by Brian Clegg, on pages 186 and 187, he notes the concept of "mathematical beauty" as a criterion that many would use when a physical theory cannot yet be tested by experiment, when considering the merits of a purely mathematical solution. But we really don't know if "mathematical beauty" is a good convention, even if it could be universally defined.
I would consider simplicity to be an important contribution by conventionalism to science, and perhaps mathematical beauty is related.
Cheers - Jim
Dear professor Tucci,
you seem to place the contributions of Mach, Duhem and Poincaré to conventionalist analysis on an equal footing with one another. Given that experience cannot impose any falsification on conventional truths, I don't understand exactly how conventionalism contributes to science in your opinion. For example, does conventionalism embrace ways of putting forward a logical framework to deal with physics? Does it only admit simple and pragmatic frameworks? Or is it related to the linking of facts to concepts?
- Badges
- Science topic
- Similar topics
- Philosophy
- Philosophy Of Science