The book is an introduction to the philosophy of the human sciences based on case studies of achievements in human sciences. The case studies focus upon researches dealing with language, action and mind.
What image of science does the book try to convey? How does it characterize the procedures of human sciences? The answer to such questions in a summary must be short and suggestive, possibly through the use of metaphors. As the natural source of metaphors is literature, two particular genres, science fiction and detective fiction, are of peculiar interest as sources of images of science and of metaphors of scientific research. For the human sciences, the detective story is of special interest. A character, in one of Agatha Christie's novels, says about the detective: “You would have made a good archaeologist, M. Poirot. You have the gift of re-creating the past.”1 The comparison may be turned around as the archaeologist also is a sort of detective; her work of inquiring resembles a detective investigation.
Philosophical investigation may also turn out to resemble the work of a detective. The present book analyzes in this style the case of the human sciences. What kind of research in social studies or arts and humanities could be qualified as scientific? The book examines first some of the usual suspects: linguistic, economic and psychological studies. Then it tries to draw its conclusions.
As for the image of science in detective fiction, one fascinating example is Arthur Conan Doyle's The Hound of the Baskervilles. Holmes and Watson adhere to what they see as the norms of scientific investigation. The image of science is that of the Nineteenth Century, but with some distinctive qualifications. Fact gathering and theorizing are separated. 2 However, Watson is puzzled; he wants to know what kinds of facts is he supposed to report. Holmes' answer draws the attention on the significance of the research questions. 3 As for Holmes' methods they are supposed to be pretty clear, but Watson never gets them fully. 4 Holmes wants to know the facts, but he does not rely on induction; he often makes deductions. A theory may well be formulated before there is enough evidence in its favor. 5
Now, even if we admit that an investigation has been carried out according to the rules of science this does not make it automatically part of science. Being like (a scientific investigation) does not mean being the same. Let us say that a detective wants to expose the evil deeds of a gang. The members of the gang communicate using a code. Breaking the code is an achievement like the decipherment of Linear B. However, the result has a different status. The outcome of a scientific research is integrated in a network of research outcomes. In the case of Linear B, the outcome is significant for Ancient Greek studies and Indo-European studies. 6 While it does not make much sense to repeat the process of the decipherment of Linear B as such, reading any new tablet written in Linear B is an indirect repetition of the decipherment. In contrast to the decipherment of Linear B, the result obtained by the detective is not integrated into a network of research outcomes; it might be useful in similar cases, it might even be used as evidence in favor of some approach in forensic sciences. But, like Sherlock Holmes, the real detective solves a case that is added to the history of crime detection.
Research outcomes are very important and the present book is focused on outcomes. Unlike Karl Popper, who denies the existence of stable outcomes, the present book assumes the existence of real outcomes, not mere “surmise and conjecture”, as Sherlock Holmes would say. 7 For a while, Ventris' theory of Linear B was a surmise and it made sense to ask for the specification of empirical data that might falsify it. But, after tests and improvement, its status changed; the research lead to an important outcome, which modified the very network of relevant research outcomes.
As it is well known, Karl Popper challenges the scientific status of the theories of Marx, Freud and Adler. He writes about Marx's theory of history that “in some of its earlier formulations (for example in Marx's analysis of the character of the 'coming social revolution') their predictions were testable, and in fact falsified” [Conjectures and Refutations (London: Routledge, 1989), p.37]. The theories of Freud and Adler, however: “were in a different class. They were simply non-testable, irrefutable” (ibidem). Despite the reference only to Marx, Freud and Adler, Popper's challenge is a broader one and has to be answered in the philosophy of the human sciences.
From the point of view of this book, Popper's challenge to the theories of human sciences requires a distinction between little and big theories. The theories of Marx or Freud attempt to be big theories, while Ventris' theory of Linear B is a little theory. The big theories in human sciences tend to become grand narratives and fail for rather different reasons than those of Popper. In other contexts, Popper's own approach to the human sciences is different. He recognizes the absence in their case of general laws and proposes a specific logical schema for explanations [see Objective Knowledge (Oxford: Clarendon Press, 1979), pp.354--355]. He is, however, tempted by what might be called a “heavy industry approach” to the human sciences.
The book also rejects Lakatos's research programmes and Kuhn's para\-digms as “heavy industry approaches” to the human sciences. In sum, from the perspective of the present book, human sciences are “light industry”.
The book proposes a replacement, in the philosophy of human sciences, for concepts such as the paradigms or the research programmes, the research keys. Research keys, as the hard core of a research program, are not directly affected by empirical evidence. They are, however, more flexible. They may be combined on different “rings” with various other research keys. Sometimes, they survive, while other keys from the ring perish. 8
Now, in order to obtain an outcome, the research must follow some procedure. Is there a method for getting a scientific outcome? The book endorses Feyerabend's anything goes, but in a weak form. All that matters is to find a scientific outcome; any method would do fine, but it has to respect certain conditions and constraints. The most important conditions are the public access to the methods of data collection and the procedures used for obtaining the outcome and the repeatability of the research outcome. An obvious constraint prevents the specification of the outcome, before the research process, in the method used for obtaining it.
The book has a double critical stance. On one hand, it rejects the “heavy industrial” approach to the human sciences. On the other hand, it rejects “interpretation” as a scientific method, specific to the human sciences. It criticizes Dilthey and the hermeneutic approach, as well as the influence of Heideggerian phenomenology in the human sciences.
The book makes a distinction between scientific research and its superstructure. A first layer of the superstructure has a narrative character. Other layers include normative extensions of the narrative superstructure. The layers of the superstructure cannot be inferred from scientific research. The proper place of interpretation is not at the level of scientific research; it plays the role of a sort of recursion between scientific research and its superstructure.
Scientific research requires openness to research methods and repeatability of outcomes; otherwise science cannot be a critical endeavor. The book endorses Merton's formula of science as “organized scepticism”[Social Theory and Social Structure (New York: Macmillan, 1968), p.601], but in a modified form. The book makes use of the distinction between organizations and institutions. Organizations have aims and plans. Institutions are systems of rules. From the perspective of the book, science is institutionalized scepticism. Scepticism is not in the aims or the plans of scientific organizations. Researchers may even have no idea about the scepticism cultivated by science and they may simply follow the procedures of their own organization; they might very well ignore the rules of science as an institution. Scepticism is the ingredient of the rules of science as an institution, not the rule of the day-to-day behavior of scientists.
Human sciences do not differ in kind from the natural sciences. There is a continuum between the natural and the human sciences, but items at the far ends of the continuum may have many differences among them and maybe only common features with the closer neighbors.
1 Agatha Christie, Murder in Mesopotamia (London: Collins, 1936), chapter 28.
2 Holmes says to Watson: “I wish you simply to report facts in the fullest possible manner to me, and you can leave me to do the theorizing.” (Arthur Conan Doyle, The Hound of the Baskervilles [London: George Newnes, 1902], p.105.)
3 “Anything which may seem to have a bearing, however indirect, upon the case” (ibidem).
4 See, for example, Watson's attempt to characterize doctor James Mortimer (The Hound of the Baskervilles, first chapter).
5 Holmes formulates the theory that Stapleton has killed old Charles Baskerville and Selden, the convict, but he resists Watson's suggestion to arrest Stapleton at once. Watson is convinced that they “have a case”. Holmes rejects this opinion: “Not a shadow of one — only surmise and conjecture. We should be laughed out of court if we came with such a story and such evidence.” (The Hound of the Baskervilles, pp.285--286.)
6 The decipherment of Linear B is discussed here in section 2.4.
7 The Hound of the Baskervilles, p.286.
8 The example analyzed in the book is that of the keys of Propp's research of the fairy tales. For a while, they had been added to the structuralist ring of research keys; however, after the demise of structuralism, Propp's research keys survived and even spread on other research areas than the original one. The details of this discussion are in section 5.4.3.