First Principles in Science: Their Epistemic Status and Justification

Breadcrumb Navigation



Day 1 (10 June, 2016)

09:15 - 09:30 Registration
09:30 - 09:45 Welcome
09:45 - 10:45 Michael Stöltzner: "Axioms as First Principles: The Case of Mathematical Physics"
Chair: Catherine Herfeld 
10:45 - 11:15 Coffee Break
11:15 - 12:00 Milena Ivanova: "The Evolution of Constitutive Principles"
Chair: Catherine Herfeld
12:00 - 12:45 Sam Fletcher: "The Principle of Stability"
Chair: Catherine Herfeld 
12:45 - 13:45 Lunch (on your own)
13:45 - 14:30 Matteo Colombo: "Explanatory Pluralism: an Unrewarding Prediction Error for Free Energy Theorists"
Chair: Matt Farr 
14:30 - 15:30 Liz Irvine: "Measurement and Introspection"
Chair: Matt Farr 
15:30 - 16:00 Coffee Break
16:00 - 16:45 Catherine Stinson: "Grounding Inferences from Model Organisms"
Chair: Seamus Bradley 
16:45 - 17:45 Samir Okasha: "Evolutionary Theory and the Strategy of Endogenization: On the Role of First Principles in Evolutionary Biology"
Chair: Seamus Bradley 
18:30 Workshop Dinner (Georgenhof)

Day 2 (11 June, 2016)

10:00 - 11:00 Kevin Hoover: "First Principles, Fallibilism, and Economics"
Chair: Catherine Herfeld 
11:00 - 11:30 Coffee Break
11:30 - 12:15 Alexander Linsbichler: "Justifying the Fundamental Axiom of Austrian Economics – What Rothbard Could Have Done, But Didn’t Do"
Chair: Milena Ivanova 
12:15 - 13:00 Paul Teller: "Evaluating First Principles on the Basis of What?"
Chair: Milena Ivanova 
13:00 - 14:00 Lunch (on your own)
14:00 - 14:45 Marco Giovanelli: "'Prinzipienfuchser'. Historical-Philosophical Considerations on Einstein's 'Principle-Strategy'"
Chair: Sam Fletcher 
14:45 - 15:30 Attila Grandpierre: "The Significance of First Principles of Science"
Chair: Sam Fletcher 
15:30 - 16:00 Coffee Break
16:00 - 17:00 Robin Hendry: "Elements and (First) Principles in Early Modern Chemistry"
Chair: Milena Ivanova 
17:00 - 17:30 General Discussion and Closing


Matteo Colombo: Explanatory Pluralism: An Unrewarding Prediction Error for Free Energy Theorists (joint work with Cory Wright)

Courtesy of its free energy formulation, the hierarchical predictive processing theory of the brain (PTB) is often claimed to be a grand unifying theory. To test this claim, we examine a central case: activity of mesocorticolimbic dopaminergic (DA) systems. After reviewing the three most prominent hypotheses of DA activity-the anhedonia, incentive salience, and reward prediction error hypotheses-we conclude that the evidence currently vindicates explanatory pluralism. This vindication implies that the grand unifying claims of advocates of PTB are unwarranted. More generally, we suggest that the form of scientific progress in the cognitive sciences is unlikely to be a single overarching grand unifying theory.

Sam Fletcher: The Principle of Stability

How can inferences from idealized models to the phenomena they represent be justified when those models deliberately distort the phenomena? Pierre Duhem considered just this problem in part II, chapter III ("Mathematical Deduction and Physical Theory") of The Aim and Structure of Physical Theory (1914), arguing that inferences and explanations from mathematical models of phenomena to real physical applications must also be demonstrated to be approximately correct when the (idealized) assumptions of the model are only approximately true. Despite being included in Duhem's most influential contribution to philosophy of science, this chapter and the principle it contains is little discussed among philosophers. Yet mathematicians and physicists both contemporaneous with and subsequent to Duhem took up this challenge (if only sometimes implicitly), yielding a novel and rich mathematical theory of stability. My goals in this presentation are thus twofold: first, to trace some of the history of this principle of stability and its precursors in reference to their application in science, and second, to present a modern version of the principle, exploring some of its applications and implications, as well as comparing it to related notions that have received more

Marco Giovanelli: 'Prinzipienfuchser'. Historical-Philosophical Considerations on Einstein's 'Principle-Strategy'

In 1919, in a brief article for the London Times, Einstein famously declared relativity theory to be a 'principle theory', like thermodynamics, rather than a 'constructive theory', like the kinetic theory of gases. Einstein later identified himself jokingly as a Prinzipienfuchser, a 'principle-fanatic'. Considerable attention has been drawn recently to such a distinction in recent literature. This paper argues that Einstein's 'principle-strategy' is based on a simple and rather widespread argumentative structure which can be traced back at least to Helmholtz. In both special relativity and thermodynamics systematical the failure to observe an expected result allowed by the given laws of physics (perpetual motion, ether drift etc.) grows from a curiosity into a non-existence claim and finally into a mathematical criterion that the laws of nature have to satisfy, if the expected effect should not happen (energy conservation, Lorentz invariance etc.). Einstein's achievement was probably not have introduced this strategy, but for having brought it to unprecedented successes. The paper concludes that successful principles in physics have often a negative or interdictory aspect. Instead of being a by-product of the actual laws of nature principles, principles are justified in as much they effectively restrict the domain of possible laws of

Attila Granpierre: The Significance of the First Principles of Science

Although the concept of the 'first principle of science' is of central importance, it suffers from basic ambiguities. Overviewing its history, we show that theoretical physics has a remarkable architecture. Observable physical phenomena are governed by the fundamental physical laws, which can be derived from the least action principle, the first principle of physics. Allowing that the basic structure of fundamental scientific theories rests upon first principles, the road becomes open to reveal the first principle of biology and psychology. We show that Bauer's principle can be formulated as the greatest action principle. It plays the same role in theoretical biology than the least action principle in physics. We obtained an unexpectedly elegant and simple unification of physics and biology on the basis of the action principle. We show that not only matter and life, but consciousness can have a first principle as well, one that also can be expressed in terms of the action principle. We point out that reduction to first principles has a fundamental significance for physics, biology and psychology, as well as for obtaining a comprehensive scientific world picture based on the sciences of matter, life and

Robin Hendry: Elements and (First) Principles in Early Modern Chemistry

Modern chemistry seems to be committed to two groups of what one might call ‘first principles’: theoretical commitments which provide a framework for detailed empirically based theories, but are not themselves directly testable. One group concerns the structures of chemical substances at the molecular scale, how these are related to their individuation as chemical substances, and the physical space in which such structures are assumed to be embedded. Another group concerns elemental composition, and the assumption that the chemical elements are actually present in their compounds, contra Aristotle. (Note that this second group emerged much earlier in the history of science.)

From an epistemological point of view, the two cases are importantly different. In the first, spatial intuition provides a framework for thinking about structures and the differences between them, so it is plausible to think that at least part of the basis of the structural first principles is a priori. In the second case (elemental composition), this seems to be lacking: Aristotle provides an alternative account of elemental composition which seems to be neither incoherent, nor ruled out by any more general framework of thought, such as geometrical

Kevin D. Hoover: First Principles, Fallibilism and Economics

First principles are most clearly displayed in pure deductive systems. Empirical disciplines require some sort of falsifiability. Fallibilism is the doctrine that no empirical beliefs can be certified as true beyond possibility of doubt. Yet, empirical inquiries also require a starting place. Even rejecting foundationalism, any inquiry starts with indubitable beliefs – that is, beliefs that are not in fact doubted. Anti-foundationalism and fallibilism are in tension, as indubitability appears to have necessary consequences, undercutting fallibilism, while fallibilism threatens confidence in the de facto first principles that begin inquiry. This tension is examined in the different attempts to define economics and its method from John Stuart Mill’s economics as the science of wealth through Lionel Robbins’ and recent neoclassical economists’ economics as the optimal use of scare resources. Commitment to first principles risks emptying economics of its empirical content, while commitment to empirical content entails violating supposed first principles and muddles the boundaries of the discipline. Just as final truth, even if not attainable, can serve as regulatory ideal, principles too can serve as a target, never to be reached. Empiricism requires that we constantly seek to resolve the tension between fallibilism and indubitability, knowing that there is no stable

Liz Irvine: Measurement and Introspection

One of the ongoing and foundational questions in psychology and cognitive science is just what role, if any, introspection is supposed to play in developing and testing theories. I will argue that while the methodological problems associated with introspection often seem unique, there are many structural similarities with problems of measurement more generally. Other phenomena are regularly affected by processes of measurement, and other measures routinely face problems of reliability, accuracy, and scope. Drawing on different approaches to measurement, I will illustrate and evaluate possible ways of moving forward in debates about how to gather and use introspective reports, though point to a pessimistic

Milena Ivanova: The Evolution of Constitutive Principles

The applicability of non-Euclidean geometries in physics and the developments of the special theory of relativity challenged the epistemic status the Euclidean axioms and the laws of mechanics had received in the Kantian framework. First steps in accounting for the choice of first principles and their revisability were made with Poincare’s development of conventionalism and Reichenbach’s relativized notion of a priori. The main aim was to account for the choice in selecting the first principles and explain theory change in cumulative and continuous way. According to the predominant contemporary account, advanced by Michael Friedman, principles are to be regarded as relativized a priori; they are taken to play a constitutive role in a theory, making it empirically meaningful, but are dynamical and change in theory transitions. This account allows for principles to be dynamic and relative to a specific framework, defending the rationality of theory transitions.
In this talk I explore the origins and developments of the relativized a priori and the conceptual problems that motivated it. I focus particularly on C.I.Lewis’s and Arthur Pap’s less studied defence of the pragmatic a priori and their accounts of the structure of scientific theories. I compare these accounts to Friedman’s and explore which framework has the resources to accommodate an interdisciplinary perspective on the function of first principles in science.

Samir Okasha: Evolutionary Theory and the Strategy of Endogenization: On the Role of First Principles

This talk concerns "first principles" of evolutionary biology, understood to mean the core principle(s) of Darwinian evolution. I focus on the question of how much in evolutionary biology can be explained in terms of these principles alone, and how much requires extraneous empirical assumptions. To focus the issue, I describe an intellectual trend in evolutionary theory over the last 150 years, which I call the 'strategy of endogenization'. This strategy refers to a particular way of increasing the generality of evolutionary theory, by devising adaptive explanations for biological features that were originally part of the background conditions against which such explanations took place. Where successful, the features in question cease to be unexplained background conditions and are brought within the fold of evolutionary theory, or endogenized. Drawing on a range of examples from evolutionary biology, historical and contemporary, I explore the logic of the strategy of endogenization and ask what its limits

Alexander Linsbichler: Justifying the Fundamental Axiom of Austrian Economics – What Rothbard Could Have Done, But Didn’t Do

The epistemological status of first principles of human behavior, which economic theories are supposedly based on, has long been fiercely discussed. In these debates, proponents of the Neo- Austrian School of economics, such as Ludwig Mises and Murray Rothbard, defend a particularly radical position. According to them, the methodology of the theoretical social sciences is praxeology. It is maintained that, since a praxeological theory of human action rests on the so-called fundamental axiom, empirical data can neither confirm nor falsify it. While the received view interprets Mises’ position as synthetic apriorism, new readings classify his epistemological position as a form of conventionalism. By contrast, Rothbard defends an essentialist justification of the fundamental axiom.
Furthermore, Rothbard formulates four requirements the fundamental axiom has to meet. It must be empirically meaningful, a priori to complex historical events, absolutely true, and not conceivably falsifiable. The paper explicates these challenges and contends the following:

  • Rothbard’s essentialist defence of praxeology does not meet his self-imposed requirements.
  • Conventionalist defences of praxeology, which Rothbard rejects, fulfil his conditions for the fundamental axiom.

Moreover, conventionalist foundations may promote acceptability of and interest in Neo-Austrian methodology with mainstream philosophers of science, social scientists, and

Catherine Stinson: Grounding Inferences from Model Organisms

In biology one of the few principles that are taken for granted, is that evolutionary history is the best indicator of biological function. This assumption is embedded in our theorizing about model organisms. That mice and humans shared an ancestor 75 million years ago justifies learning about humans by doing experiments on mice. I point out several problems with this assumption, and offer an alternative account of inference from model organisms.

The problems are that the choice of model organism clearly involves pragmatic factors, generalizations are usually made to species not clades, and the common ancestors that hypothetically ground these inferences may be completely unknown, or not exist at all. Alternative justifications for expecting similarity across species include living under similar conditions, having faced similar evolutionary challenges, or filling similar niches.

What matters is how well the model approximates the target kind, and how robust that kind is in terms of the mechanisms of interest. Good model organisms can be tightly controlled, so as to exclude as many incidental factors as possible. There is an epistemic (not merely pragmatic) benefit to using models with well-developed experimental technologies, like mice or fruit flies. Mechanisms that have been retained down hereditary lines are often robust, but it is this robustness, not common ancestry, that grounds inferences between related

Michael Stöltzner: Axioms as First Principles: The Case of Mathematical Physics

Physics is the discipline most influential for philosophers’ understanding of first principles. Core predictions of our most basic theories – or thus the received view has it – can largely be deduced from first principles and supplementary conditions that select the physical models and provide boundary values. Properly mathematized and laid down in suitable axioms, first principles and their application conditions can then be logically interpreted and philosophically analyzed. Recent debates about models and explanations have shown that even the best first principles of physics hardly ever reach down to the concrete applications. Axiomatized quantum field theories have been unsuccessful in integrating the standard model of elementary particle physics. Does this show that even philosophers of physics have to root for first principles that are not mathematical structures, but closer to the concrete facts? My paper argues that there remains still space for mathematical axioms as first principles if one does not overtax mathematical physics and equate it to the first step of logical analysis. Instead my paper suggests to take inspiration from the factual achievements of mathematical physics – both in the days of Hilbert and Poincaré and in the renaissance of mathematical physics after 1960 – and understand axioms in a more explorative or opportunistic fashion. Such a view is also closer to recent developments in the philosophy of mathematics and brings the first principles of physics again closer to other scientific

Paul Teller: Evaluate Frist Principles on the Basis of What?

Choice of first principles must be sensitive to their function. Classically their function is as basic source from which we may derive more specific truths. As such truth was a primary requirement. But in all the sciences, science gives us collections of models that are never both completely precise and completely accurate. This circumstance leads Giere to recommend that we think of first principles as model building guidelines, in analogy to basic dress patterns. As such first principles need to be sensitive not only as guides to accurate representations but as guides to the kind of information in which we are particularly interested and guides to formulations that are humanly usable in practice. With this revised perspective we also revisit the question of whether nonetheless, and if so how, first principles themselves contain information about the world. The short paper will not attempt concrete solutions but will aim to provide a prolegomenon to these issues by regimenting them as clearly as possible.