Husserl and the Sciences

Richard A. Feist, Richard A. Feist

• Publisher: University of Ottawa Press
• Year: 2004
• ISBN:
  9780776618661
• Topic: Filosofía
• Subtopic: Fenomenología en filosofía

Edmund Husserl (1859-1938) is one of the previous century's most important thinkers. Often regarded as the "Father of phenomenology, " this collection of essays reveals that he is indeed much more than that. The breadth of Husserl's thought is considerable and much remains unexplored. An underlying theme of this volume is that Husserl is constantly returning to origins, revising his thought in the light of new knowledge offered by the sciences.

PART I

PHENOMENOLOGY, EPISTEMOLOGY, AND THE SCIENCES

CHAPTER ONE

EDMUND HUSSERL AND THE HISTORY OF CLASSICAL FOUNDATIONALISM

Herman Philipse

1. INTRODUCTION: PROBLEMS AND METHOD

According to many present-day epistemologists, the justification of scientific theories is relative in at least two respects. Whether a specific theory is justified at time t depends both upon the set E of empirical data available at t, and upon the set R of rival theories which are considered by the relevant scientific community at that time. Indeed, a theory is justified at time t if and only if it performs better than its rivals in terms of the accepted criteria for theory choice, and one decisive criterion for theory choice is some version of the criterion of empirical superiority. As a consequence, a theory that is now justified might cease to be justified in the future because of two reasons. New empirical data may top the balance in favour of an existing rival theory, or a new rival theory may be designed that performs better.

This relative notion of the justification of theories by competition is not only a rough model of justification, it also solves the traditional problem of demarcation. We do not need to look for an atemporal criterion of demarcation that enables us to decide whether theories are scientific or not (such as the criterion of falsifiability) for we now have a time-relative criterion. Theories are not scientific or unscientific in the abstract, but someone has an unscientific attitude if he or she prefers a given theory T at time t even though at that time there is another theory available that is clearly superior to T in terms of the criteria for theory choice.¹

The model of justification by competition must be applied also at the meta-level of justifying epistemological theories. Like scientific theories, epistemological theories and models are not invented and evaluated in the abstract. They are typically designed to solve problems that arise in determinate historical circumstances, for example because new developments in the sciences conflict with the epistemological status quo. The criteria for theory choice in the domain of empirical science have their analogues in the domain of epistemology, although here, of course, the data set does not consist of natural phenomena; rather, it is the set K of (scientific) knowledge at a given time t. Normative epistemological theories should be empirically adequate in the sense that they enable us to provide a rational reconstruction of paradigmatic scientific developments. An epistemological theory, then, is justified at time t if it provides a rational reconstruction of the scientific developments up to t that is superior to the reconstructions provided by its rivals. This meta-epistemology of epistemologies explains the co-evolution of science and epistemology.

In the tradition of analytical philosophy, the rules of academic etiquette seem to exclude an application of this model of justification by competition on the meta-level, since the model prescribes that in trying to understand and evaluate epistemologies, specific historical circumstances must be taken into account. Analytical philosophers, however, typically assume that it is imperative to discuss the major epistemological models proposed in the past, such as foundationalism and coherentism, without reference to the historical contexts in which these models were developed as if the evaluation of epistemologies could proceed in an intellectual vacuum.² In my view, this a-historical style of analytical epistemology accounts for both the characteristic barrenness of the discipline and its irrelevance to scientists and the history of science. Moreover, in many cases analytical presentations of epistemological models such as foundationalism are distorted, because authors abstract from the time-bound reasons for developing these models in specific historical circumstances, and omit to discuss features that were once deemed essential.

In this paper I want to explore by means of a case study how these defects may be remedied, focusing mainly on the example of Edmund Husserl’s foundationalism. I shall attempt to answer three questions. First, which type of foundationalism did Husserl adhere to? Second, was he justified at the time in endorsing his type of foundationalism as a general theory of science? Finally, would one currently be justified in accepting Husserl’s foundationalism? I argue that Husserl’s foundationalism is a version of what might be called ‘classical foundationalism,’ so named because it ultimately derives from Aristotle. A succinct rational reconstruction of the history of epistemology from Aristotle to Husserl will show that classical foundationalism is a research programme whose development was determined not only by an internal logic but also by the scientific context. During Husserl’s lifetime, the programme degenerated, and at the end of this paper I consider a section of Heidegger’s Sein und Zeit which shows that the programme could not cope with the scientific revolutions of the early twentieth century.³

2. THE ORIGINS OF CLASSICAL FOUNDATIONALISM

In the literature there are various definitions of foundationalism, either as a theory of knowledge or as a theory of justification. These definitions have two points in common. All varieties of foundationalism require a distinction, among beliefs that count as justified or as knowledge, between those that are basic and those that are derived. Furthermore, foundationalist theories hold that justification is predominantly one-directional, that is, from basic beliefs to derived beliefs, and that justification of non-basic beliefs is a logical derivation of some kind.

Within the broad class of foundationalist theories, so defined, many different types of foundationalism may be distinguished, depending upon the answers that these theories provide to three questions:

1) which kinds of beliefs are considered to be ‘basic’?

2) how are these basic beliefs justified?

3) how is the logic of derivation to be characterized?

Sometimes, the honorary title of ‘classical foundationalism’ is reserved for the doctrine that basic beliefs are beliefs that concern the nature of our own sensory states, our own immediate experience.⁴ But from a historical point of view, this terminology is unfortunate. Both British empiricism and some logical positivists who held this view represent stages in the development of a much older research programme in epistemology, a programme inaugurated by Plato and Aristotle. Instead, the epitheton ornans ‘classical’ should be linked to a feature common to all or most phases in the development of this programme. These phases may then be called ‘classical’ because they were in accordance with the ancient model of foundationalism constructed by Aristotle.⁵

Within the context of Greek philosophy, foundationalism was developed as an answer to two problems that were pressing at the time: (a) how is knowledge (episteme) to be characterized in contradistinction to mere opinion (doxa), and (b) how can one show that knowledge is possible at all? Plato’s main dialogue on this issue, Theaetetus, ends in an aporia with regard to (a). Socrates had argued that knowledge should not be defined as mere true judgment, because judgments made upon hearsay cannot count as knowledge even when they are true. With his characteristic irony, Plato now makes Theaetetus suddenly remember (201c/d) what he once heard a man say, to wit, that knowledge is true judgment with an account logos. But this theory leads into insuperable difficulties. Having an account might mean, for instance, that one knows an object O if one is able to go through its elements. This interpretation either leads to a regressus ad infinitum (in order to know the elements, one has to go through the elements of these elements, etc.) or one should conclude that one cannot know the elements, which is absurd, for how could one know the whole without knowing the elements? Having an account might also mean being able to tell the difference between the object O and other objects. This second interpretation makes the definition of knowledge circular, for how could one tell the difference if one does not know this difference?

We may suppose that these final sections of Theaetetus inspired Aristotle’s epistemological research programme. In Prior Analytics, Aristotle had developed his deductive logic. On the basis of his theory of syllogistics, he could explain in Posterior Analytics what it is to have an ‘account’ logos. In chapter A2, Aristotle defines “knowledge” as a true judgment with an account, and he explains what it is to have an account by saying that an account is a scientific deduction of the true judgment, that is, a deductive proof the premises of which are explanatory with regard to the conclusion (71b, 17–19). It may seem that the requirement of deductive proof is extravagant because we now think that it cannot be met in the empirical sciences. Yet the requirement is plausible given Aristotle’s common-sensical approach to knowledge. It is part of the logical grammar of the verb ‘to know’ (and its Greek counterparts) that propositions of the form ‘X knows that p’ imply that p is true; this is why truth is made into a requirement for knowledge. And if X can have no knowledge unless his or her opinion is true, it is plausible to assume that X’s claim to knowledge must be warranted by an account which guarantees the truth of what X claims to know, that is, by a proof.⁶

Aristotle’s definition of knowledge as belief or judgment proven to be true by an explanatory deduction answers question (a), but it immediately raises the sceptical question (b): how can one show that knowledge as defined is possible at all? In particular, how is one able to avoid the objections of regression and circularity which Socrates raised in Theaetetus? It is no accident that Aristotle discusses these objections in chapter A3 of Posterior Analytics, and he proposes what I shall call ‘classical foundationalism’ as an answer to these objections. The problem with which Aristotle deals here has been baptized, instructively, the Münchhausen trilemma.⁷ If knowledge that p is belief that p, proven to be true, we cannot know that p unless we also know that the premises of the proof that p are true, which we cannot know unless we also know that the premises of the proofs that the premises of the proof that p are true, are true, and so on. Aristotle admits that knowledge would be impossible if this regress goes on indefinitely, “for it is impossible to go through indefinitely many things.”⁸ We might avoid the regress by admitting circular deductions, but these do not amount to proofs: they beg the question. Aristotle’s solution consists in claiming that there are basic beliefs or “first principles” that are known to be true without proof. Accordingly, the definition of knowledge as belief proven to be true cannot apply to them, and Aristotle calls knowledge of the first principles sofia instead of episteme. A deductivist variety of foundationalism is the result of these considerations, and it is this deductivist variety which I call ‘classical foundationalism.’

Classical foundationalism is best conceived as a research programme, for although it provides a definite answer to 3) by characterizing the mode of derivation of derived knowledge as deductive proof, it leaves open issues 1) which kinds of beliefs are considered to be ‘basic’ and 2) how these basic beliefs are justified. Let me call the complex issue (1 & 2) the ‘problem of the first principles’ in classical foundationalism. The history of Western epistemology from Aristotle to Husserl and Heidegger may be reconstructed, to a large extent, as the development of the research programme of classical foundationalism which has been the dominating epistemological programme for more than two millennia. Like his younger contemporary Euclid, Aristotle distinguished two types of first principles: principles common to all the sciences and principles specific to a particular discipline, such as astronomy or physics. It would be the task of philosophy to lay the foundations of all the special sciences by establishing the common first principles, and this is why Aristotle (and Descartes) used the expression ‘first philosophy.’ Of course philosophy also had to solve problem 2) both for these general principles and for the principles of the special sciences. Aristotle solved this latter problem in chapter B19 of Posterior Analytics.

In order to understand Aristotle’s solution, one must grasp the specific form that the problem of the (specific) first principles acquired within the framework of classical foundationalism. In the sciences, we have to deduce general theories from first principles. Aristotle held that ‘basic’ knowledge in the empirical sciences must be justified by observation, and that observation is of individuals. Yet universal theories cannot be deduced from singular judgments of observation. Hence, the first principles of the sciences must be universal too, and Aristotle claimed that the logical form of scientific proofs is what the scholastics called Barbara. Moreover, scientific proofs cannot establish the truth of theories unless our knowledge of the first principles is secure. For this reason, Aristotle required that the first principles are necessarily true and that we are able to grasp these necessary truths. The problem which Aristotle had to solve may now be formulated as a paradox: classical (deductivist) foundationalism seems to exclude empiricism, because empiricism appears to imply that the first principles of the sciences are singular judgments of observation, whereas classical foundationalism requires universal first principles that are necessarily true. How are we to combine classical foundationalism with empiricism? How can observation justify first principles that are necessarily true and universal?

This is the problem Aristotle set out to solve in chapter B19 of Posterior Analytics, and he solved it by...