Philosophical and Foundational Problems of Modern Physics

Network Project Proposal approved by the European Science Foundation


Abstract

This project proposal requests financial support for a 3 year Network entitled Philosophical and Foundational Problems of Modern Physics. The aim of the Network is to facilitate and intensify European cooperation in the philosophy and foundations of modern physics. This will be done by creating a framework that links European institutions specializing in this interdisciplinary field. The main planned activities of the Network are: organizing three conferences, supporting some smaller­scale workshops at European institutions and creating and maintaining a web site to serve as an efficient platform for the exchange and dissemination of information.
     The present project proposal

     • describes the background, motivation and aims of the Network (Section 1);
     • specifies the activities planned (Section 2);
     • concisely describes the three subfields in philosophy of physics to which the
       conferences are devoted (Sections 3, 4 and 5);
     • gives the names of the members of the coordinating committee (Section 6).

1 Background, rationale and aim of project

The emergence of modern physical theories in the second half of the 19th and in the first half of 20th century has had a profound effect on philosophy: new physical theories forced one to re­think many old philosophical concepts and problems. The influence went also the other way: physics was in need of an analysis of its changing conceptual foundations, and philosophy contributed substantially and positively, by not only commenting from a distance, but by clarifying and scrutinizing foundational issues in a way that has led to new physics.
     To a large extent, modern philosophy of science was born out of this mutually fertilizing intellectual interaction. Many scientists who played a leading role in the scientific revolutions of the 20th century, among them Einstein, Bohr, Heisenberg and Schrödinger, took an active part in discussions about the conceptual foundations of the new physics. As Hans Reichenbach, a prominent 20th century philosopher of science, observes in his The Rise of Scientific Philosophy, it was soon realized that one cannot do full justice to the philosophical and conceptual problems of modern science by providing footnote­like comments on them. The modern professional philosopher of science, trained in logic, in (some of) the sciences and producing systematic analyses of particular sciences, is the result of this recognition. So is the field of philosophy of science itself: during the 20th century philosophy of science, and philosophy of physics in particular, have established themselves as major academic disciplines with increasingly distinct academic communities and institutions.
     Parallel to the three main fundamental modern physical theories, statistical mechanics, quantum theory and theory of relativity, philosophy of physics itself can be divided into three subfields, each centered around the conceptual and interpretational problems relating to these three theories. Sections 3, 4 and 5 of this project proposal contain concise statements describing the major problems currently discussed in these respective subfields in philosophy of physics.
     The aim of the planned Network is to create a framework that facilitates research and the dissemination of results in the field of philosophy of modern physics, concentrating on these three subfields. The main activity planned by the Network is the organization of three conferences and a number of related small­scale activities at European institutions; this will be backed up by the creation of a platform for the exchange and dissemination of information (see Section 2 for more details about the planned activities).
     Several interpretational problems cut across the boundaries of the three disciplines in question, making these subfields inter­dependent. More generally, a characteristic feature of modern philosophy of science, and of the philosophy of modern physics in particular, is its interdisciplinary nature: the problems and their analyses combine elements from physics, mathematics, logic, philosophy and the history of science. This entails that, to be successful, research in this field needs constant communication between researchers from different disciplines, in particular between philosophers and physicists. Therefore, special attention will be paid to the creation of possibilities for interaction between experts from different disciplines. The Network will thus provide a platform for multi­disciplinary discussion.
     The conscious effort to get physicists involved is also motivated by another consideration. The study of the conceptual foundations of physics leads to the creation of a stimulating intellectual climate that is conducive to high­quality research in physics. Accordingly, the activities of the planned Network are not meant to be restricted to the circle of philosophers of physics, but are intended to have an impact also in the much larger academic community of physicists. Achieving this goal will become more likely if active, leading physicists participate in the Network (see Section 2).
     The three modern physical theories that the Network's activities concentrate upon were born in Europe, and the conceptual and philosophical investigations of these theories were also begun in Europe. European (or European­born) physicists and philosophers of physics dominated the field until about the 1960s. But during the second half of the 20th century, study of the foundations and philosophy of physics became ever stronger in the U.S.A., and a very well­trained younger generation entered the field in this period. The academic work of this generation was helped tremendously by university departments and research centers specializing in philosophy of science and by institutions such as the U.S.A.'s Philososophy of Science Association (PSA) and National Science Foundation (NSF). The PSA has been organizing nationwide PSA conferences every two years for the past 30 years, and publishes Philosophy of Science, a leading journal. The PSA conferences and the journal have been important vehicles to foster research and disseminate results. NSF has also been crucial in the development of philosophy and foundations of physics by offering grants to support research in this field. (See Section 6 for plans to get U.S. scholars involved in the Network.)
     No similar Europe­wide institutions have stimulated the philosophy of physics, or even of science in general, in Europe during the same period. This is certainly one of the reasons why Europe has lost its dominance in the philosophy of physics in the past 40 years. Yet Europe has managed to remain competitive, and a number of European scholars have distinguished themselves as international leaders in the discipline. Europe has a few university departments whose faculty include these international leaders, and where teaching of philosophy of physics is done on an internationally competitive level (Oxford and London, U.K.; Utrecht, The Netherlands; Budapest, Hungary; Florence, Italy). Another sign of the strength of Europe is the very successful launch in 1995 in Cambridge, England of a new journal, Studies in the History and Philosophy of Modern Physics.
     Most scholars of the foundations and philosophy of physics in Europe are employed, however, by philosophy or physics departments that do not have the resources needed to run a special training programme and are thus especially in need of contacts, cooperation and communication. Creating the Network will help to maintain Europe's competitiveness in philosophy and foundations of modern physics and thereby to maintain a rich and important tradition in Europe's intellectual life.

2 Activities planned

The Network aims at organizing three sorts of activities

• Three major conferences in the discipline
• A number of small workshops
• Creating an electronic platform for the exchange and dissemination of information

Conferences

     Three conferences are planned with the following tentative titles and locations

1. Philosophical and Foundational Issues in Statistical Mechanics, in Utrecht, The Netherlands
2. Philosophical and Foundational Issues in Space­time Theories, in Oxford, U.K.
3. Philosophical and Foundational Issues in Quantum Theories, in Budapest, Hungary

     These meetings are intended to bring together European researchers from the several relevant disciplines. The conferences should also contribute to the training of young researchers. Therefore, we propose that: (i) experts from inside and outside Europe be invited to participate and present overviews; (ii) registration fees be kept to a minimum; (iii) we might select the best papers that are not already forthcoming in a learned journal, to be published in a conference volume; and (iv) that financial support be given to young scholars and researchers from Eastern Europe in order to encourage their participation.
     Sections 3, 4 and 5 contain concise statements describing the major problems in the respective subfields. The typical conference programme will consist of invited review talks by physicists and philosophers, contributed talks, and short talks. The short talks are mainly intended to provide opportunity for young scholars and graduate students to present their results.

Small workshops

     About one fifth of the budget will be earmarked for the organization of small workshops and for stimulating visits to European institutes by international experts. The purpose of the workshops is to intensify cooperation between European institutions. These meetings will be devoted to in­depth study of specialized subjects. The Coordinating Committee will encourage that working physicists also get invited to give talks addressing conceptual and foundational issues.

Exchange and dissemination of information

     We consider it vital that there will be an efficient flow of information within the Network. We intend to create a professional Web­site for this purpose, where detailed programmes of all activities will be published, where institutes can post announcements, course materials and similar texts, and through which contacts can be established. In addition, electronic mailings will be used to disseminate news.

3 Philosophical and Foundational Issues in Statistical Mechanics

Statistical theories in physics emerged in the second half of the nineteenth century in the pioneering work of Maxwell, Boltzmann and Gibbs. The aim of these authors was to provide an explanation of thermal phenomena in terms of an underlying mechanical reality. Particularly famous is Boltzmann's statistical­mechanical explanation of irreversible processes, governed by the Second Law of Thermodynamics.
     However, conceptual and philosophical problems soon surfaced, in particular problems surrounding the famous "reversibility objections'', the nature and meaning of probability and ensembles, the status and justification of the Ergodic Hypothesis, the assumption of Molecular Chaos, etc.
     In the philosophical literature, the study of such problems developed into standard accounts in the period 1950­1970, in the work of Reichenbach, Popper, Grünbaum, Gold and others. Subsequently, interest in philosophy of statistical physics waned
for several decades, partly because the fundamental problems of quantum mechanics attracted much attention, and because of the hope that a full solution of the foundational problems of the latter theory might also solve (at least some) of the foundational problems of statistical physics.      Recently, however, the philosophy of statistical physics has been enjoying a veritable come­back to the center­stage of the philosophy of physics, as witnessed by several philosophical monographs. There are several reasons for this renewed interest in philosophical issues of statistical mechanics. First, there is a growing awareness that, in spite of the standard accounts mentioned above, the central issues are still very much open. Secondly, it has become clear that although some interpretations of quantum mechanics do throw new light on the problems of statistical physics, no unproblematic solution to these old problems is provided by quantum mechanics. But perhaps even more importantly, the physical literature has in the meantime shown great progress in the development of new approaches to statistical physics, using sophisticated methods of analysis, extending far beyond the original methods of Maxwell, Boltzmann and Gibbs. (To list a few: the work of Lanford, the so­called "BBGKY''­approach, the "rigorous results'' approach of Ruelle et alia, Prigogine, the "subjective'' approach of Jaynes, modern developments in chaos theory and ergodic theory, and the renormalisation group approach to phase transitions.) Also the formulation of the classical theory of thermodynamics has improved greatly in recent axiomatic approaches, such as those of Lieb and Yngvason. Thus it has now become imperative to take these modern developments into account in any philosophical analysis.
     To be specific, the main issues which we propose to address in this Network, and in the planned conference, in particular include the following:

• The statistical explanation of the appearance of irreversible processes and the nature of entropy.
• The problem whether thermodynamics can be reduced to (some form of) statistical mechanics.
• To classify and assess the various schools and approaches within modern statistical mechanics with respect to the question whether and how they accomplish the above goals.
• The role played by the various interpretations of probability in the statistical assumptions employed in such approaches, and the different possibilities they offer.

4 Philosophical and Foundational Issues in Space­time Theories

Problems of space and time have traditionally taken center­stage in work on the conceptual foundations of physics, and of science in general. Already before the beginning of modern science in the 17th century the topic was discussed in ways that are still relevant for us today, although the answers that we can give have become very different from the old ones. Are space and time independent entities, "substances'' that exist independently of material objects; or are they systems of relations between objects? Is space finite or infinite? Can there be a beginning of time? What explains the various apparent asymmetries between past and future, e.g. that causes precede their effects and that there are traces of the past but not of the future?
     In the context of modern physics, the first question---whether space is an independent substance or a system of relations---became famous through the exchange between Newton and Leibniz, and the later contributions to the debate by Mach and Einstein. Originally, Einstein thought he had vindicated a Leibnizian or Machian relationist point of view with his general theory of relativity. Later analyses, however, argued that the general theory of relativity, far from justifying the relationist viewpoint, supported substantivalism. It was argued, for example, that the existence of space­time points has to be presupposed before the field equations of general relativity can even be formulated. But the issue is still very much open; in particular because in the last fifteen years the so­called "hole­argument'' has demonstrated that substantivalism leads to a radical form of indeterminism in general relativity.
     The general theory of relativity encompasses cosmological models, and so offers a new framework for the old questions of whether the universe is spatially finite and of whether it has a beginning in time. It has become clear that solutions of the field equations as a rule exhibit singularities, and the usual point of view is that the beginning of our universe can be represented by such a singularity (the Big Bang). Does this mean that it is meaningless to speak about time before the Big Bang (the standard viewpoint)?
     The question has not been decided yet, and there are proposals for schemes in which it does make sense to speak about times before the Big Bang, or in which there are no singularities at all.
     Concerning time, questions can be asked that are very similar to those about space. Is time independent, progressing on its own (as Newton thought); or is it rather a system of relations between events, such as changes in material objects? Additional questions, specific to time, pertain to its asymmetry. The basis for the pervasive temporal asymmetries of the processes occurring around us is still unclear, in spite of the great amount of effort devoted to it. The essential problem is that the fundamental laws of physics are time­symmetrical, so that time­asymmetries cannot be derived from them without making some time­asymmetrical assumptions. This topic is intimately connected with questions in the foundations of statistical physics (Section 3).
     Although these traditional issues in the foundations of space­time theories are still open, much progress has been made in their study. This is partly because new developments in physics have made it possible to formulate the problems in new and sharper ways, and have suggested new solutions.
     Developments in physics has also led to new questions. For example, philosophers of physics now focus much attention on the role of symmetry principles (space­time symmetries, like Lorentz symmetry and reflection symmetry; but also gauge symmetries). Also new questions, in connection with black holes and other space­time singularities, unusual topologies, and closed time­like curves, have emerged. Finally, questions relating to the combination of gravitational theory and quantum mechanics, and so­called "Theories of Everything'', have begun to draw the attention of workers interested in foundational research.
     To be specific, the main issues which we propose to address in this Network, and in the planned conference, in particular include the following:

• The recent debate about substantivalism, in particular in the light of the hole­argument.
• Philosophical issues about cosmology, including e.g. the initial singularity, inflationary cosmology and the cosmological constant problem.
• Philosophical aspects of symmetries, including gauge­symmetries in a non­quantum context.
• Philosophical issues raised by singularities and pathological topologies, for example in black holes.

5 Philosophical and Foundational Issues in Quantum Theories

Ever since quantum theory was born in 1900, it has been beset with interpretative difficulties. We shall pick out five main areas, which continue to be areas of active research. During the 1920s, quantum theory took a definite shape as a result of intense work by physicists and mathematicians; and the theory reached its canonical form in von Neumann's 1932 book Mathematical Foundations of Quantum Mechanics. This book formulated sharply two major interpretational issues: (1) The problem of whether the statistical character of quantum theory is epistemic or ontic in nature (also known as the problem of "hidden variables''); and (2) the measurement problem.
     (1): While von Neumann thought he had answered (1) definitively by providing an "exact proof '' that the probabilistic character of quantum theory is irreducible and cannot be thought of as a result of our ignorance, the issue cannot be considered settled even today. One central aspect of this problem is the question how to interpret the quantum probability calculus, which differs from the classical Kolmogorovian probability calculus in crucial respects. Is there a meaningful relative frequency interpretation of a probability measure defined on the non­commutative event structure determined by quantum theory? It has been claimed that quantum probabilities are not any sort of peculiar, exotic probabilities: they are ordinary, classical conditional probabilities, conditioned by the (also classical) events that certain measurements are carried out. Is this a viable re­interpretation of quantum probability? If so, what philosophical consequences does it entail?
     (2): The measurement problem arises in quantum theory if one attempts to describe the measuring process on the assumption that both the measured system and the measuring apparatus are quantum systems and so described by quantum theory. The problem is that accepting the so­called "eigenvalue eigenstate link'' rule, which specifies the conditions under which a quantity for a quantum system has a definite value, leads to a paradoxical situation: after the measurement interaction the joint system consisting of the measured+measuring system ends up in a joint quantum state in which the measuring apparatus does not have a sharp pointer value, i.e. does not in fact provide a definite result of the measurement. Over the years since von Neumann's book, many solutions of the measurement problem have been proposed: for example, appeals to the environment of the measuring apparatus (decoherence), to "other worlds'' (i.e. Everettian interpretations), or to hidden variables, i.e. extra values for quantities. (This last appeal can be made in various ways: one can postulate once and for all that certain quantities, say positions, are definite, as in the pilot­wave interpretation; or one can let which quantity is definite depend on the quantum state, as in the modal interpretation.) But none of these solutions has won assent; controversy still rages.
     (3): Soon after von Neumann's book, the Einstein­Podolsky­Rosen (EPR) paper of 1935, and later in the 1960s, the closely related work of J. Bell articulated another area of interpretative difficulty: non­locality. The question here is whether certain probabilistic correlations between spacelike separated events predicted by quantum theory (and confirmed by experiments) can have a causal explanation. Again, the subject has flourished since these ground­breaking papers. For example, in 1989 Greenberger­Horne­Zeilinger (GHZ) exhibited a new type of situation in which the quantum correlations violate classical ideas in a non­probabilistic way. More philosophically, many specifications of the notion of causal explanation have been proposed and analyzed; most recently, using the idea of a branching space­time to represent indeterminism.
     (4): Quantum logic was originally proposed in 1936 by G. Birkhoff and von Neumann: the idea was that the lattice structure determined by the projections of a quantum system should be interpreted as the logic of propositions about the system. What notion of logic does this involve? In particular, does it mean that logic is empirical? Since 1936, quantum logic has developed into an extremely rich field, which is still expanding in several directions. Recently, research has turned to the investigation of more abstract algebraic structures (e.g. quantum MV algebras, unsharp quantum logic), and connections have also been made between quantum logic and category theory.
     Discussion of the conceptual status and philosophical significance of these recent directions of research is much needed.
     (5): Quantum field theories, especially relativistic ones (QFT), have been developed since about 1930. There are many different approaches to these; but foundational and philosophical attention has naturally focussed on mathematically rigorous approaches, especially the algebraic relativistic quantum field theories (ARQFT) developed since the 1950s. QFTs present us with new interpretative issues, in addition to those above, which are common to all quantum theories. Two such issues are as follows. (i): The notion of a particle. Various no­go results (many based on the Reeh­Schlieder theorem) seem to entail that the notion of particle as sharply localized in space or spacetime is incompatible with the basic assumptions of QFT. What then is the notion of particle in QFT? And more generally: What ontology is QFT compatible with? These questions have been addressed, for example in the modal interpretation; but many questions remain open. (ii): The QFT version of EPR­Bell correlations. In the last fifteen years, it has been shown that various kinds of non­locality - more precisely, entanglement between local algebras of observables pertaining to space­like separated space­time regions - are generic in AQFT. Thus the problem arises whether one can causally explain these correlations. Here again, many natural problems are entirely open.
     To be specific, the main issues which we propose to address in this Network, and in the planned conference, include the following:

• Recent approaches to the measurement problem; including decoherence, and Everettian, pilot­wave and modal interpretations.
• Recent approaches to quantum non­locality; including branching space­time approaches, and synchronization models.
• The philosophical assessment of recent developments in quantum logic, such as quantum MV algebras and unsharp quantum logic.
• The interpretative issues posed by QFTs.

6 Members of the Coordinating Committee

Names of members (in alphabetical order):