Dilthey on the Concept of Psychology—I


This is the second of a set of seven course notes written in 1998 for a seven week introduction to the thought of Wilhelm Dilthey. It deals with Chapters One and Two of his Ideas concerning a Descriptive and Analytic Psychology

Dilthey’s Ideas concerning a Descriptive and Analytic Psychology—Chapters One and Two

Some Preliminary Observations

  1. If one wants to see a kind of psychology which Dilthey would regard as explanative and based on incorrect assumptions about the nature of the human psyché, I suggest Jerry Fodor’s “Methodological Solipsism Considered as a Research Strategy in Cognitive Psychology”, in Dreyfus, Hubert, and Hall, Harrison, (eds.), Husserl, Intentionality and Cognitive Science, Cambridge, Mass.: M.I.T. Press, 1982, pp.277-303. This article illustrates very well the kind of faculty psychology which Dilthey dislikes. It also well illustrates the ‘synthetic’ approach which Dilthey attributes to explanative psychology, namely, the identification of elementary psychological entities, in Fodor’s case, mental representations and modules for processing the same and then the synthesis of these elements and modules into a merely causally interrelated whole which is supposed to provide a model of such and such a psychological process, e.g., cognition, perception and the like.

  2. For two instructive accounts of naturalism, see the articles on naturalism in The Blackwell Companion to Metaphysics and the Encyclopaedia of Philosophy.

  3. In these notes, I occasionally refer to Dilthey’s text simply as Ideas.

Dilthey’s Ideas concerning a Descriptive and Analytic Psychology—Chapters One and Two

It is common to divide Dilthey’s views on the need for, and nature of, a grounding of the so-called human studies into two periods, a ‘psychological’ one extending through the 1880’s up to just before the turn of the century, and a more ‘hermeneutic’ one from, say, 1900 onwards. This distinction appears to me to be at best artificial although it is certainly true that around 1900 or slightly before Dilthey did move from talking about the fundamental method and approach of the human studies as descriptively psychological to hermeneutic. This reflects, I think, an appreciation on Dilthey’s part that by and large the human studies investigate domains of human reality which, due to differences in culture and tradition, or perhaps the passage of time, are no longer readily accessible. Thus, it is wrong to assume that the kind of activity which the student of human affair engages in when studying different or past cultures is just an extension to new domains of an ability to understand others, their actions and their works which we exercise in everyday life. Because wherever there is distance and difference, human reality is not effortlessly intelligible, the understanding displayed by the human studies is not just a reflective, more explicit version of what we do in unproblematic everyday contexts. A genuinely hermeneutic effort is needed; the people, the cultures or the acts one is studying are so to speak broken texts whose meaning is not directly accessible; it must rather be discerned in the vehicles which express them in a manner not strictly comparable to how we effortlessly understand one another in everyday contexts. Talk of the method and approach of the human studies as descriptive obscures this difference entirely.

At the same time, it would be wrong to exaggerate the difference between Dilthey’s so-called ‘psychological’ and his ‘hermeneutic’ periods. Much is preserved, in particular, his central terminoloy of the ‘psychological’ or ‘psychical nexus’ (psychischer Zusammenhang) and the holistic account of socio-cultural reality that goes with this term of art. In the next set of notes, I will start unpacking some of the things Dilthey says in the first two chapters of his “Ideas concerning a Descriptive and Analytic Psychology.” This text was written in 1894 and thus belongs to his ‘psychological’ period. The first chapter is entitled “The Task of a Psychological Grounding of the Human Studies”, the second, “The Distinction between Explanative and Descriptive Psychology”. In particular, I want to concentrate on pp.138-150 of Chapter I and the whole of chapter II, which is quite short, namely, pp.154-158. In order, however, to understand properly what Dilthey is on about in these opening chapters, a few rough observations on the history of psychology are necessary.

Very roughly, one can distinguish two strands and attitudes within contemporary psychology. On the one side there is, to appropriate a distinction from William James, a tender-minded kind of psychology: this is the kind of psychology we associate with the names of Freud, Jung, Binswanger, Klein, Fromm, R.D. Laing and numerous others. It is often clinically and therapeutically oriented and in its theoretical accounts of its clinical and therapeutic methods it often, if not always and not always unambiguously, spurns any suggestion that it is a natural science at all, much less a natural science which could take physics as its model. On the other side there is the tough-minded kind of psychology, e.g., cognitive psychology, some brands of development psychology, the psychology of perception, and so on. This kind of psychology very often bumptiously proclaims its status as a natural science. Some prominent recent and contemporary representatives are Karmiloff-Smith, Smolenksy, Rumelhart, McClelland and many others.

We need to keep these two strands strictly apart in attempting to understand Dilthey. Although it has many roots and sources in the tradition of Romantic and philosophical psychology which was eventually eclipsed by the hard-nosed variety, the tender-minded variety of psychology strand is actually the younger. For it arose out of traditions of opposition to the general conceptions underlying the natural scientifically oriented brand. This latter brand in fact arose during Dilthey’s life, and it is this brand of psychology which Dilthey has in mind when he speaks of ‘erklärende Psychologie’, i.e., ‘explanatory’ or ‘explanative’ psychology.

In Dilthey’s time, psychology in this natural scientifically oriented sense was just starting to assert itself. One of the first people to articulate the idea of a natural scientifically oriented ‘explanatory’ psychology was Kant´s successor at Königsberg, Johann Friedrich Herbart (1776-1841). Although he insisted that in psychology one still needed a number of indispensable metaphysical presuppositions, in particular, in order to account for the unity of psychological phenomena and the particular way they change, he insisted that one could have a ‘physics of the mind’ („Physik des Geistes“) which utilised mathematical and quantitative techniques, thereby enably psychology to constitute itself as a genuine science. Herbart´s followers, e.g., Waitz (1821-1864) und Drobisch (1802-1896),1 In these notes I was unable to discuss the third fundamental characteristic, namely, the human propensity to regard certain kinds of action as not just wise or unwise, but morally right or wrong.

elaborated on these ideas unlike by the mid-century numerous thinkers were claiming that psychology had to be done as an independent, empirical discipline, without any debt to, or baggage from, traditional philosophy and metaphysics. Mid-nineteenth century scientists, doctors and philosophers like Wundt (1832-1920), Helmholtz and Fechner (1801-1887) set about establishing a psychology whose model was physics and which conceived its object of study as neither made of immaterial substance, nor identical with the human soul. Psychology thus conceived was to be neutral about these kinds of issue, although of course the more it succeeded in cognising psychological phenomena without having to talk about immaterial substances or of souls, the less plausible it made these doctrines. (Something similar applies in the case of physics and God.)

Dilthey identifies a number of contemporary traditions and individual thinkers as the first advocates of this idea of an independent, natural scientifically oriented psychology which takes physics as its model, hence similarly aims to exploit experimental and mathematical techniques. In this connection, he mentions 1) the tradition of associationist psychology, which stems from the British empiricists Locke, Berkeley and Hume; 2) Herbart (1776-1841), who has already been mentioned; 3) Friedrich Albert Lange (1828-1875), the early Neokantian author of the very influential book Geschichte des Materialismus (The History of Materialism); and 4) the English psychologist Herbert Spencer and Taine. A natural scientifically oriented psychology conceives its task as that of identifying the constituent elements, causal laws and mechanisms of psychological phenomena, just as physics identifies constituent elements, causal laws and mechanisms of physical phenomena. Identification of the fundamental elements of things and of how these fundamental elements interact causally to produce certain kinds of phenomenon is what Dilthey means by the explanation of such phenomena. A science which aims to explain in this sense is precisely an explanatory or explanative science. As Dilthey puts it, “(u)nter einer erklärenden Wissenschaft ist jede Unterordnung eines Erscheinungsgebietes unter einen Kausalzusammenhang vermittels einer begrenzten Zahl von eindeutig bestimmten Elementen (d.h. Bestandteilen des Zusammenhangs) zu verstehen.” (S.139) It should be noted that while such explanative science and the explanations it comes up with may very well and perhaps even ideally take mathematical form, this is not necessary. Many natural scientific explanations, from contemporary explanations of the origin of the universe to genetic explanations of hereditary disease are strictly qualitative, even though they may involve much calculation around the edges.

Dilthey points out that explanative psychology in this sense works essentially with hypotheses. Insofar as to form a hypothesis is to formulate a general proposition hopefully valid for all cases on the basis of a limited number of cases, there is nothing to object to this. But, says Dilthey, the formulation of hypotheses has acquired a more definite, specific sense in natural scientific contexts, and thus in all disciplines which, like explanative psychology, seek to model themselves on natural sciences. According to Dilthey, we do not experience the causal interconnectedness of things, but only their temporal succession. When a flame causes water to boil, I see the flame burning in close proximity to the kettle, and I also see the water boiling in the kettle. I do not, however, see in addition to these two events any further event of the flame’s causing the water to boil. In order to establish that a causal connection exists between these two events, I must test out the hypothesis that such a connection exists by putting kettles on flames a number of times. I will then have a highly probable degree of certainty that a causal connection exists. (S.140-141)

But, ask Dilthey, can this approach be transferred across to our psychological life? (S.142) That the representatives of an explanative psychology do intend such a transferral, and that there are a number of points at which this idea is problematic, are things Dilthey wants to show in the course of this work. We can, or so Dilthey seems to suggest, get a better preliminary understanding of what it means thus to conceive psychology on the model of natural science, and where this idea goes wrong by appreciating the following: in contemporary explanative psychology we do indeed find, as a matter of fact, lots of hypotheses. Unfortunately, this is an embarrassment of riches: “Ein Kampf aller gegen alle tobt auf ihrem Gebiete nicht minder heftig als auf dem Felde der Metaphysik. Noch ist nirgends am fernsten Horizonte etwas sichtbar, was diesen Kampf zu entscheiden die Kraft haben möchte.” (S.142) Dilthey seems to be voicing about contemporary explanative psychology much the same sentiment that Kant voiced about traditional metaphysics: “… (s)o far … are the students of metaphysics from exhibiting any kind of unanimity in their contentions, that metaphysics has rather to be regarded as a battle-ground … in which no participant has ever yet succeeded in gaining even so much as an inch of territory, not at least in such manner as to secure him in its permanent possession.”2

Given the subsequent development of explanative psychology, indeed, of any kind of psychology, Dilthey is making a quite valid point. Since the rise of explanative and experimental psychology in the late nineteenth century, there have been a succession of attempts to establish psychology on the same kind of sure footing which physics, in particular through Sir Isaac Newton, was able to establish itself. Newton’s mechanics provided a research framework for future scientists which lasted approximately 250 years. And in a certain sense Newton’s work has not been superseded, even today: most standard engineering and technological work utilises classical mechanics, indeed, classical mechanics are often used in the design of experimental apparatus and instruments used in theoretical physics of a quite high-powered kind. Finally, many people regard Newton’s theory as just a special case of which Einstein provided the more general theory.

Nothing like this has occurred in psychology. Initially, in the late nineteenth century there was associationist psychology which derived from classical empiricism and was often highly introspectionist in nature. That is, it was quite happy to maintain that psychological data and even laws could be discovered by the psychologist’s observing how he or she experienced things. Such observation of one’s own mental life was known as ‘inner perception’, something of which Dilthey himself makes much mention. But in the late nineteenth century, psychologists began calling introspection into question, denying that it provided the kind of empirical data a natural scientifically oriented psychology could profitably use. The point was not, or not just, that the observations one made about what went on in one’s own breast were no more reliable and infallible than ordinary observation of external things. The point was rather that not matter how reliable or infallible such data were, they were of limited use in experimental, explanative psychology. The aim of such a psychology is to establish the various mechanisms and lawful connections whereby physical things out there cause psychological phenomena ‘in the mind’; and conversely, things ‘in the mind’ cause things to happen in the physical world. About these mechanisms and laws very little knowledge is gained by introspection. What is needed is experimentation from a third-person, non-introspectionist point of view—experimentation involving diverse test subjects which would seek to correlate their introspective reports of their psychological states and events with observations made by the experimenter of various physical and physiological facts, events and changes.

Eventually, this hostility to introspection went so far that one did not merely deny that it could establish anything of scientific interest. One increasingly began to declare that the phenomena allegedly introspected did not really exist. At least from the standpoint of the natural scientifically oriented psychologist one could regard the entities talked about in introspective reports, e.g., colour sensations, feelings, perceptions, beliefs, desires, intentions and the like as entities postulated by a bad or false theory, and thus with precisely the same status as witches and fairies. This hostility to what had traditionally been understood by psychological phenomena gather pace during the late nineteenth century, culminating in the first decades of the twentieth century in the movement known as behaviourism. (The canonical text of this movement was J.B. Watson’s book Behaviourism, which appeared in 1912.) This was a kind of experimental, natural scientific psychology which tried to do without any talk of mental or psychological entites, except in a behavioural sense. If beliefs, desires, intentions and the like existed at all, then only as dispositions to behave: to say that I believe that it is raining is simply to say that I will tend to do such things as put on my raincoat, open up an umbrella, or put my newspaper over my head, etc., before I go outside.

Although behaviorism received some competition from the so-called Gestalt psychologists, it was the dominant school of psychology at least in Anglo-american countries for many decades. But by the fifties it was quite clear to most people that behaviourism was not the promised equivalent to Newtonian mechanics, that foundation which would allow a scientific psychology to march relentless forward for hundreds of years. So people began looking for a completely new approach. By the late fifties, if not earlier, psychologists and philosophers of psychology believed they have found the new approach. They began to re-invent the mental—but only because the model of the computer suggested how one could acknowledge the mental ad important elements in a scientific psychology without lapsing back into unscientific introspectionism. Thinking was like the running of a computer programme, beliefs like the stored information used by the programme and desires the various objectives the programme served. The task of scientific psychology was thus to come up with models of the programmes we humans run when thinking and acting. Psychology thus conceived is known as functionalism, or more accurately, as computationalism, which is one way of spelling out the idea that beliefs, desires, perceptions are mental functions, i.e., neuro-physiological configurations with certain properties to change, and be changed by, other such configurations. (Analogously, when a computer is running a certain programme, its hardware is configured to respond and interact in certain ways.)

Unfortunately, various problems with this conception have arisen. The computationalist attempt to spell out the general functionalist idea has not delivered the long-awaited foundation for the sure march forward of psychological and in particular, of cognitive science, which is of course the psychology of cognition and perception. Enthusiasm for it is thus waning and now the latest fashion is the idea of neural networks and, more latterly, the idea that the human psyché can be modelled scientifically by conceiving it as a non-linear dynamical system.3

So Dilthey is quite right, indeed prescient, in his observation that there is in psychology a situation not unlike that in what it particularly despises, namely, metaphysics. That there is indeed this situation is indeed something of which many natural scientifically oriented psychologists show themselves to be painfully aware. Just like so many economists, many psychologists protest over and over again that what they are doing is ‘science’. Contrast this to the much more relaxed attitude of many contemporary physicists, who are rightly quite sure that physics, whatever it is, is a huge success, who thus do not feel any need to defend the status of physics as a science (or anything else).

Dilthey gives his own example of a fundamental hypothesis which he regards as subject to endless, undecidable controversy. This is appears to be the standard reductive materialist hypothesis that psychological facts and events are caused in law-like ways by physical and physiological facts and events, and are indeed reducible to these latter. (I take it by speaking of “die Zurückführung aller Bewußtseinserscheinungen auf atomartig vorgestellte Elemente, welche in gesetzlichen Verhältnissen auf einander wirken …” (S.142-143), Dilthey does intend not merely an epiphenomenalism, but a genuine reductionism.)

Dilthey points out that proponents of explanative psychology like to defend their making of such hypotheses by appeal to the natural sciences, by which Dilthey of course means physics and chemistry (see S.139). But I think he is trying to say that this begs the crucial question of whether psychology may in fact such natural sciences as its model. If the answer to this question is ‘Yes’, then certainly this characteristic defence is sufficient; one may indeed respond that, as the experience and history of the natural sciences show, we may have every confidence that the current controversial character of this hypothesis will, by further work and experiment, be overcome. But until this question has been answered affirmatively, one may just as legitimately take the long-standing controversial character of such hypotheses as indicating that something is wrong in one’s conception of what psychology is and thus what it may take as its model.

At any rate, Dilthey (S.143) insists here on a point deriving from Aristotle4 that no discipline at all, much less any human study, should take its methodological lead from any other established discipline; rather, it needs to adopt the most general and obvious methodological strictures which apply to any discipline, e.g., the need to have appropriate reasons for what one claims, and work out what these general strictures come to in its own particular case, i.e., in view of its particular subject matter or universe of discourse. This is what needs to be done in order to set up any new discipline satisfactorily. Although Dilthey does not intimate anything along these lines here, I think there is already a gesture here to Dilthey’s fundamental project in the philosophy of the human studies, a project which Dilthey himself once called the Critique of Historical Reason.5 This project is precisely to investigate critically just what is involved in engaging in such human studies as psychology, history, sociology and the like. One should not just assume that psychology, for example, is so like the established natural sciences that it can use the very same methods and assumptions, and thus become a natural science itself. One has to ask the recognisably Kantian question of how psychology, history, sociology and the like are possible as ‘sciences’, that is, as genuinely systematic, disciplined forms of theoretical inquiry. When Kant asked how physics and mathematics are possible as sciences, he did not doubt that they were thus possible; he was not interested, at least not primarily, in solving any sceptical worries about natural and mathematical science. He wanted to determine just what such science is, that is, how the practice of natural scientific and mathematical inquiry is able to live up to its pretensions to provide genuine theoretical knowledge of a physical or mathematical reality independent of it. Dilthey extends this idea to the up and coming human studies or Geisteswissenschaften. How are they possible as sciences? That they are possible is something he certainly does not doubt. But he does want a philosophical account of how they are possible, an account which determines the nature of the psychological, historical and/or social realities these disciplines study, and on this basis works out what their appropriate methods are and are not.

Now according to Dilthey one obvious difference between the human studies and the natural sciences is that the latter investigate entities which, as Dilthey puts it, “… appear in consciousness as coming from outside, as phenomena and as given singly ….” (S.143) The human studies, on the other hand, investigate entities which appear as internal, as reality and as embedded in a living web, as interconnected with other such entities. What on Earth does Dilthey mean here? His principal point comes out on the next page, where he says that in the case of the human studies the experienced nexus or matrix or web of entities is what comes first, the distinguishing of individual elements in the web comes afterwards (see S.144). Evidently, the point is this: the entities investigated in the human studies are essentially parts of wholes and are studied as such.6 The human studies approach their objects of inquiry as entities that have certain internal, identity-defining relations to other similarly embedded, situated entities—whether these entities be mental states and experiences, which always and only occur as parts of the psychological life-process (Seelenleben) of an individual human being; or whether they be individual human beings, works of art, social institutions and the like, which always and only occur as parts of a larger social, cultural and historical life-process. Indeed, as we shall shortly see, the entities studied by the Geisteswissenschaften are not merely logically embedded in a whole, in the way in which the poles of a magnet, for example, are logically embedded in the magnet itself. Rather, the relevant whole is a dynamic, temporally extended, indeed developmental process in which causal relations exist between the individual entities in the whole and the whole itself. So the entities studied by the Geisteswissenschaften are logically and causally embedded in their respective whole. All this is common to the different kinds of entity studied by the different kinds of Geisteswissenschaft.

This holistic embeddedness is, I take it, what motivates Dilthey’s repeated use of the word ‘Zusammenhang’, which occurs for the first time on S.143, where Dilthey says that the entities studied by the Geisteswissenschaften are given to the investigating consciousness “… als ein lebendiger Zusammenhang ….” (S.143) The word ‘Zusammenhang’ is one of Dilthey’s favourite words; it occurs combined in all sorts of compounds, e.g., ‘seelischer Zusammenhang’, ‘Lebenszusammenhang’, etc. In the English translation this important word is translated by the word ‘nexus’. This translation is not ideal, for the word ‘nexus’ is an obscure foreign word from Latin,7 whereas as the German is a much more everyday word, hence is more graphic and colourful. It seems to me that at least in some of the contexts in which Dilthey uses the word ‘Zusammenhang’ one could have just as well used the more graphic English words ‘web’, ‘matrix’ or ‘network’. But given the translator’s preference for the world ‘nexus’, the various compounds Dilthey makes out of the German word end up being translated as, e.g., ‘psychological nexus’ (‘seelischer Zusammenhang’), and ‘life-nexus’ (‘Lebenszusammenhang’), and so on.

Of some importance is Dilthey’s notion of the ‘acquired psychological nexus’ (‘erworbenener seelischer Zusammenhang’), of which he speaks on S.144. If in general, a ‘Zusammenhang’ is an interconnection, a web or matrix of certain kinds of entity which logically precedes and causally shapes these entities, then it is fairly clear what Dilthey means by the ‘acquired psychological nexus’. Each individual human qua psychological being has, or rather is, a certain acquired psychological nexus. The acquired psychological nexus is the whole life-process of the individual—life-process not in a biological or physiological sense, but in the sense of the psychological process of human living which an individual at any given moment in their biological lives is and which is embedded in a larger social, cultural and historical process of human Life, i.e., Life-with-a-capital-L. Because the acquired psychological nexus is embedded in a larger Life—Life-with-a-capital-L—just as the specific abilities, faculties and psychological states of an individual are embedded in the acquired psychological nexus, this larger Life is itself a ‘nexus’, the nexus of the individual’s social, cultural and historical world. It seems that in general a Zusammenhang is any dynamic, processual whole with the following characteristics:

  1. the existence of the whole is logically prior in the sense that it is presupposed by the existence of any of its parts, so that these parts are not separable components or building blocks; and

  2. the whole interacts causally with the parts, so that the whole can actually shape the individual parts just as much as these parts can shape the whole.

There are, however, specific kinds of nexus, for example, the psychological nexus which constitutes the individual human being and the overall social, cultural and historical nexus in which the human being is embedded. Evidently, ‘Zusammenhänge’ are a bit like Chinese boxes, namely, nested in one another.

Now as already intimated, the acquired psychological nexus, and indeed, through it, the larger social, cultural and historical nexus, can, thinks Dilthey, shape the conscious processes of thought, decision and action of this invidiual in unconscious ways. On p.144, for example, he speaks of two ways in which the conscious thought-processes of an individual can be causally conditioned: in the first instance, they can be shaped by something Dilthey calls ‘reproduction’, which unfortunately he does not explain any further. In the second instance, however, these conscious thought-processes can be influenced by the acquired psychological nexus, which is, as he here explicitly says, not accessible to our consciousness. There is thus a causal shaping relation between the individual’s whole life-process and individual events and processes within it. At the level of individual psychology and of psychological study, everything that the individual has become—their particular character, abilities, talents and dispositions—shapes their conscious thought-processes. By extension, at the leve of history and historical study, everything that the individual’s socio-cultural world has become through its life-process, i.e., through history, will similarly shape the individual’s thoughts, decisions and actions. Note here the way Dilthey assimilates the level of the individual and the level of society and history; these have structurally isomorphic developmental trajectories. Or, as it is sometimes put, the ontogenetic development of the individual, the individual’s relation to their own past, is strictly parallel to the phylogenetic development of the society and culture, and the way this society and culture stands to its past. This is characteristic of Dilthey and others in the nineteenth century. In particular, it is a crucial feature of Hegel’s conception of history—see, e.g., his Phänomenologie des Geistes.

Now one crucial task of the Geisteswissenschaften will of course be to show how the relevant nexus shapes the very identity of certain elements in the nexus: how the acquired psychological nexus of an individual shapes this latter’s thought and action, how the larger social context shapes the individual, and how history shapes the larger social context. Crucially, Dilthey thinks that to accomplish this task is not to form hypotheses in any natural scientific sense. The fact that the human studies, unlike the natural sciences, study entities in their capacity as holistically bound up with, or situated in, some such larger, dynamic process of development determines, says Dilthey, “… a very large difference in the methods by which we study psychological life, history and society ….” (S.144) The holistic character of the individual elements in the psychological life of an individual, or in the social life of a community, means that hypotheses cannot play the same role in psychology and the other human studies as in the natural sciences. He puts the point I think he wants to make in a very misleading way: he says that in the natural sciences, the formation of hypotheses in the process of theoretical inquiry is what creates the nexus or web, whereas in psychology as he conceives it, and thus in the Geisteswissenschaften generally, “… the nexus is primordially and continually given in lived experiencing.” (S.144) It is surely wrong to say that the activity of hypothesising in natural scientific investigation literally creates the causal interrelation of things.

What, however, I think Dilthey means is the following: the natural sciences operate with a notion of causation such that the effect is logically independent of the cause. In no sense is the causce itself somehow implicit in, or imprinted on, what is caused. So I cannot encounter the cause through the effect itself; by describing and understanding the effect, I gain no clues as to the cause. By looking at the boiling water, I will never discover what causes it to boil. I have to look at the surrounding facts and events. I might observe that while the water is boiling, a fly is walking up the wall, Ray Martin is hosting Carols by Candlelight yet again, or that a flame is burning under the kettle. I will now have to make a hypothesis as to the causal connection, and then engage in repeated observation of similar situations, each time altering some detail, until finally I am able to determine which correlation corresponds to a relation of cause and effect.

But, thinks Dilthey, in the study of human reality we are concerned with causal influences which imprint themselves upon the identity of their effects. This is why psychology, and indeed any other human study is descriptive rather than explanative. If by explanation we simply mean the identification of why this or that happened, then one will find explanation in this comparatively weak sense in the human studies just as much as in the natural sciences. But by explanation Dilthey means not just the identification of any old causes, but rather the explanation of phenomenon in terms of causes which are logically distinct from, in no way imprinted on, their effects. In natural science, thinks Dilthey, we move by the hypothetical, experimental method just illustrated by my example of the boiling water back to causes which, because they are not indicated themselves by their effects, have to be hypothetically postulated, which postulation has then to be confirmed by repeated experiment. (This is what moves Dilthey to say, quite misleadingly, that the process of hypothesis formation in natural science creates the physically causal nexus of things.)

But in the case of the human studies, we do not have to postulate essentially hidden causes, but rather to follow out the hints as to their causes offered by the effects themselves. We examine, for example, a historical figure like Schleiermacher or Goethe, and we see in their behaviour and personality the influences that made them what they are. We see, for example, in the behaviour of Immanuel Kant that he was molly-coddled by his mother. We see in the authoritarian and at times callous behaviour of Frederick the Great the effect of the cruelty his father displayed to him as a young man—a cruelty which culminated in his father’s forcing him to watch the beheading of Frederick’s childhood friend, Hans Hermann von Katte. (Both had attempted to flee Prussia for England; apparently, his father, Frederick William I, seriously considered executing Frederick as well but relented.) In other words, we describe these phenomena, not in order to come up with a boring, unconnected list of facts in opposition to any causal account of why the phenomena are as they are, but rather precisely in order to identify the appropriate causes. It is, I think, a great mistake to think that when Dilthey characterises the human studies as descriptive, he is saying that such studies are interested merely in what is normally understood by description, namely, some list of features and properties which perhaps enable us to identify a certain individual but identify no causality. The human studies avoid explanative hypotheses in the natural science in favour of description precisely in order to identify their kinds of causal relation, i.e., to ‘explain’ in the weaker sense of simply indicating how and why such and such has come to pass.

Note that this is not to say that the kinds of causal relation investigated by the human studies can simply be read off their effects. When I see or read about Frederick the Great’s severe and at times callous personality, I do not thereby learn that he was given a tough time by his father. The point is rather this: when I discover that Frederick was given a tough time, this fact makes immediately intelligible why Frederick is as he is. It seems that I discover here a cause which did not merely make some effect happen, namely, Frederick’s callous nature, but which makes it intelligible that Frederick should have this nature. Indeed, it seems I discover a cause which shapes Frederick in a certain way by making this shape an intelligible, understandable response to this cause.

This suggestion of a kind of causal relation which is not a purely external one, where the effect consists in being an intelligible response to the cause, is of course very vague. One may indeed dispute that there is any real, objective difference at all between the causal relation between Frederick’s nature and the way his father treated him, and the kind of causal relation one investigates in natural scientific study of people. Frederick could, for example, have had a poor memory, and this poor memory could have been caused by, say, chemical imbalances in his brain. This causal relation is clearly like the relation between the boiling water and the flame under the kettle: one needs to observe lots of correlations between short temper or poor memory and chemical imbalances in order to come up with a hypothesis that this correlation is one of cause and effect. One might argue, however, that the correlation between Frederick’s callous nature and the way his father treated him is not substantially different from the correlation between Frederick’s poor memory and chemical imbalances in his brain. Surely, one has to have heard or experienced lots of instances where fathers have mistreated their sons in similar ways with similar results. Certainly, it is on the basis of such straightforward observation of correlations between child abuse and having been the victim of child abuse that we know that paedophiles tend themselves to have been victims of paedophiles. Thus, whatever difference there might be in subjective ‘feel’ about these different causal relations, there is no objective difference. We do not have here any objective difference in kinds of causality.

It is not easy to refute this counterargument. Nonetheless, in a somewhat half-hearted defence of the suggestion that there is something special about the causal relation between Frederick’s callous nature and his father’s treatment of him, one might note the following: while it is true that we have found out about the connection between child abuse and having oneself been a victim by observing numerous correlations, it is not obvious that all comparable general claims in and of our everyday folk-psychological understanding of people are derived in this way. One may well suspect that our knowledge that a callous disposition can be the result of mistreatment by someone in the past is not in fact acquired this way. This is perhaps intimated by the fact that without having done any great empirical research we can certainly reproach another who mistreats either their children or their animals by saying, “You’ll make them vicious!” Where does our non-empirically derived confidence in this come from? Presumably it derives from the fact that we can imagine this causal relation simply on the basis of our understanding of what it is like to be the victim of cruelty.

This, in particular, the connection with abilities to imagine what it is like to be thus and so, is something we will have to come back to. In the meantime, however, we have enough to see why Dilthey insists that hypothesis, understood in the natural scientific sense as the postulation of hidden causes in no way implicit in their effects, has only a marginal, ancillary role in the human studies. (S.145) In particular, I think he wants to say that the kind of hypothesising which involves experiment, i.e., the artificially instigated and controlled repetition of phenomena in order so to speak to make the causal relations visible, only has a marginal role. By experiment one can certainly establish relatively local correlations between physical and psychological phenomena. One can establish, for example that when a finger is placed on a red-hot piece of iron, the C-fibres in the subject’s brain start firing and the subject feels pain. One can even establish very complex lawful correlations between, say, the angles at which light is reflected from objects and reports of these objects as having this or that colour. This is, however, a far cry from a unified, natural scientific theory of the psychological as such. Such correlations are, however complex they may be, still relative low-level, i.e, at the level of generalities like “Water boils at 100 degrees”. They certainly do not constitute a comprehensive, total psychological theory of phenomena like cognition, perception, personality, and so on.

Indeed, Dilthey regards the very fact that the explanative, experimental psychologists have not made much progress towards the ideal of a total theory of the psychological as indicating that this ideal is illusory. There has, he thinks, been no progress towards identifying the physical and physiological causal relations underly the various kinds of causal shaping and influencing which the human studies investigate: “Insbesondere die für die konstruktive Psychologie so entscheidende Frage nach den ursächlichen Verhältnissen, welche die Beeinflussung bewußter Prozesse vom erworbenen seelischen Zusammenhang her sowie der Reproduktion bedingen, ist ihrer Lösung noch um keinen Schritt näher geführt worden.” (S.145) As we have seen, one might well be able to correlate isolated physical and physiological phenomena with isolated psychological phenomena in the manner illustrated by the example of the red-hot iron, C-fibres and pain. But Dilthey thinks that there is and can be no strictly natural scientific account of the kind of psychological and historical causal relations illustrated above with reference to Kant and Frederick the Great. (To my knowledge, the developments in psychology since Dilthey by and large bear him out. Not that this has had any significant impact on the enthusiasm for a natural scientifically oriented psychology. Perhaps this is because the follow wry comment from Dilthey still remains true: “Die Vertreter einer solchen Hypothesenverbindung haben das schärfste Auge für das, was ihr zur Bestätigung dient, und sie sind ganz blind für das, was ihr widerspricht.” (S.145))

But if, as Dilthey alleges, attempts at an explanative, natural scientifically oriented psychology have not been successful, perhaps the appropriate thing to do is not to persist, simply insisting that one day we will get there. Perhaps the appropriate thing to do is to ask whether another conception of psychology’s procedure and method could avoid founding our understanding of psychological phenomena on a bunch of conflicting and theoretically impotent hypotheses. “So legen wir uns die Frage vor, ob nicht ein anderes Verfahren in der Psychologie … die Fundierung unseres Verständnisses von allem Seelenleben auf einen Inbegriff von Hypothesen vermeiden könne.” (S.145) Dilthey will eventually call this alternative conception descriptive and analytic psychology.

Dilthey makes clear why he thinks we need to formulate quite explicitly such an alternative conception. Thus far, investigators concerned to establish psychology as organised, disciplined process of genuinely theoretical inquiry have felt they had to model psychology on the natural sciences because it seemed to them that they had no alternative. Unless they accepted this model, they felt that psychology could never be more than a matter of writing up and codifying what analytical philosophers these days call ‘folk-psychology’, that is, our everyday pre-theoretical discourse about people’s beliefs, desires, plans, emotions and character dispositions. This concern not to rely on native intuitions, on “what everybody knows”, about human nature, is perfectly justified. But unfortunately no alternative to natural scientific quantification and experimentation has been available, and so such investigators have attempted to transfer natural scientific criteria and methods for determining the legitimacy or otherwise of theoretical claims across to psychological studies.

This has had, thinks Dilthey, a pernicious effect on the whole of the human studies. Dilthey is convinced that psychology is a human study, and in one trivial sense he is of course right: psychology studies human psychological phenomena. But he is also convinced that the very nature of psychological phenomena is such that one cannot effectively or substantially employ with them the methods appropriate for physical phenomena. As we have already seen, this has to do with the essentially holistic character of the psychological, a holism which distinguishes it from the purely physical and ranks it alongside the social, cultural and historical. In other words, this feature of the psychological makes it a human study in a stronger, non-trivial sense. It places psychology as a ‘Geisteswissenschaft’ alongside sociology, history and cultural studies in opposition to ‘Naturwissenschaft’. Here, at the bottom of S.145, a further and very important conviction of Dilthey’s manifests itself: psychology is not merely one human study amongst many, it is the foundational human study upon whose results all other human studies in some sense depend. So now Dilthey can claim that the attempt to conceive psychology on the model of the natural sciences has had a pernicious effect not merely on psychology itself, but on all the human studies, on all the ‘Geisteswissenschaften’. Because psychology is, as Dilthey believes, the foundational human study, the other human studies have seen themselves confronted by the choice of either drawing up a theoretical psychology in appropriately modelled on natural science, or of falling back on the folk-psychology of everyday life. In the first case, a natural scientifically oriented, explanative psychology transmits its inappropriate character across to the other human studies. In the latter case these other human studies find themselves incapable of establishing themselves as independent processes of theoretical inquiry governed by their own distinctive norms and criteria of theoretical legitimacy.

So in Dilthey’s eyes the time is ripe for an explicitly philosophical investigation of the very concept of psychology, one which explains just what psychology really is and ought to be, and which thereby provides the true model for the other human studies. On S.146 Dilthey indicates another benefit of such an investigation: if, as he does believe, epistemology and the theory of knowledge, requires input from psychology, then it will be crucial to the success of these kinds of inquiry that the true nature of psychology as a form of theoretical inquiry be properly understood. If psychology be conceived on the model of natural science, then this will introduce into epistemological investigations a kind of empirical dependence and hypotheticalness which is surely inappropriate for it. (S.146) So the clarification of the nature of psychology will be of use for epistemology and the theory of knowledge as much as for the human studies.

In this argument that the way to ground the human studies in their independence from all natural scientific models lies in clarifying the nature of psychology a crucial role is obviously played by the premiss that psychology is not merely one human study amongst many, it is the exemplary or paradigmatic, indeed foundational human study. Dilthey is of course aware that this premiss needs justification. Thus, on S.147 he acknowledges the need to show that all other human studies do need to draw upon and presuppose psychology if they are to avoid being either mere write-ups of everyday common sense or empirically insensitive metaphysical speculation. Here, he says, it must be shown “… daß jeder Versuch, eine Erfahrungswissenschaft des Geistes ohne Psychologie herzustellen, … unmöglich zu einem benutzbaren Ergebnis führen kann.” (S.147)

Dilthey insists that one can show of every human study that it presupposes psychological knowledge. The study of a cultural phenomenon like religion finds itself having to use concepts like feeling, will, dependence, freedom, motive and so on. These are psychological concepts, as is the notion of religious consciousness, or consciousness of God. Similarly, “(d)ie Jurisprudenz hat in Begriffen wie Norm, Gesetz, Zurechnungsfähigkeit psychische Zusammensetzungen vor sich, welche eine psychologische Analyse fordern.” (S.147) Jurisprudence cannot articulate how sentiments of natural justice and fairness operate, how goals are realised in the law or how human will is subject to the law without a clear understanding of what is involved in the psychological nexus which constitute the psychological life of an individual. Similarly, the sociological and political disciplines need to appeal to psychologically charged notions such as community, domination and dependence, while history and the study of art and literature rely on notions of the beautiful and the sublime, the humourous and the ridiculous, and so on. The literarian historian can, says Dilthey, not understand the creative life of a poet without being acquainted with the processes of artistic imagination. In general, all cultural, economic, religious, artistic and scientific systems and institutions can only be understood on the basis of an understanding of the living interconnectedness of the human soul. (S.147) All these phenomena constitute or involve different types of ‘Zusammenhang’, of nexus in Dilthey’s special sense of a holistic, dynamic, developmental process. And they do this because of the holistic, dynamic, developmental process which constitutes the psychological life and identities of the psychological beings which make them up. (S.148)

But not merely do the other human studies use psychological concepts and knowledge. Dilthey, unlike many people today, believes that different disciplines are only possible as a systematically ordered whole or totality. The human studies must form a system because the various ‘Zusammenhänge’, the individual systems and totalities which they study, themselves form a system. “Die Verbindungen, in welchen Wirtschaft, Recht, Religion, Kunst, Wissen untereinander und mit der äußeren Organisation der menschlichen Gesellschaft stehen, können doch nur aus dem umfassenden, gleichförmigen seelischen Zusammenhang verständlich gemacht werden, aus dem sie nebeneinander entsprungen sind und kraft dessen sie in jeder psychischen Lebenseinheit zusammen bestehen, ohne sich gegenseitig zu verwirren oder zu zersetzen.” (S.148) So on Dilthey’s picture there are Zusammenhänge everywhere, system and sub-system, totality and sub-totality, nested within and alongside one another. Each human studies investigates some such totality and slots into a unity with other human studies corresponding to the way the totalities studied connect up with one another.

It is not obvious what this idea of unity, both at the level of the human studies themselves and at the level of the entities studied by the human studies, has to do with psychology. Dilthey’s talk of the nexus in which the relations between the various human studies are grounded (S.148) as psychological suggest that in a manner reminiscent of Hegel before him Dilthey sees the kind of unity in which being an individual person, with a distinct personal identity, consists, as a model for the unity of a society or culture, i.e., for the way individual humans, institutions, practices, etc., are bound together in the one society, in the one culture. If this is so, then it becomes intelligible why he should insist that the human studies require insight into the structure and unity of the psychological nexus in which the distinctive identity of a person is realised. An understanding of the psychological kind of nexus enables each human study to understand just how the various nexi are unified, both in themselves, and with one another. And by understanding how the totalities they study are unified both internally and to other such totalities, the human studies are able to organise themselves into a systematic whole.

All in all, Dilthey thinks that it is a big mistake to regard the human studies as able to get by without psychological input—the right kind of psychological input, of course, not input from a psychology which takes natural science as its model. Having argued this, Dilthey moves on to discuss the relation between psychology on the one hand and epistemology, or rather the theory of knowledge (Erkenntnistheorie), on the other. In this regard, we need to note two things: firstly, Dilthey rejects the view of those who say that epistemology, or the theory of knowledge, can be done independently of psychology. Dilthey has in mind “… Kant’s school …” (S.149), i.e., the so-called Neokantians, e.g., Wilhelm Windelband (1848-1915) in South West Germany and Cohen (1842-1918) in Marburg. Each of these thinkers tried to revive Kant each in their own way. But both agreed that Kant’s work had to be freed from what they thought was a residual confusion of genuinely epistemological issue with psychology. Throughout Kant’s Critique of Pure Reason there would indeed appear to be much psychology. Kant speaks repeatedly of various faculties in the mind which apprehend and synthesis what appear to be sensations into a coherent experience of an external world. Neokantians like Windelband thought that this was just a confusion; the identification and grounding of the presuppositions of various disciplines had nothing to do with psychology of this kind.

Dilthey agrees that psychology of this kind has no role to play in the transcendental philosophical investigation of fundamental presuppositions. His remarks at the tope of S.149 indicate that Dilthey, too, is opposed to the hard divisions of Kant’s doctrine of faculties. But he associates psychology of this kind with explanative psychology. He agrees that explanative psychology has no role to play in the kind of transcendental philosophical investigation Kant envisaged. But this does not mean, as the Neokantians think, that all psychology is irrelevant; they only think this because they do not distinguish between explanative and descriptive psychology. According to Dilthey, the latter does have an indispensable role in any theory of knowledge Kantian in nature. This is shown precisely by the fact that psychological notions occur in Kant. Indeed, psychology is so indispensable for epistemology that according to Dilthey if one does not admit it consciously and in a controlled way, it asserts itself surreptitiously and uncritically. Just this, Dilthey says, happened to Kant: the basic concepts of his critique of reason certainly do derive from a definite psychological tradition. (S.149) And because Kant did not critically reflect upon the role of psychology in his work, the influence of the psychological school which impacted him is not critically controlled.8 It is because Kant does not resort to psychology in a controlled and critical way, not because he resorts to psychology at all, that Kant leaves himself open to the criticisms the Neokantians make of their master. In particular, the lack of critical control misleads Kant into formulating two important distinctions in inadequate ways.

The first distinction is that between ‘Anschauung’ and ‘Denken.’ The word ‘Denken’ means thought or thinking. The word ‘Anschauung’ is usually translated as intuition, but Kant certainly does not mean anything mystical or magical by it. What precisely intuition is according to Kant is, I think, chronically unclear. Are intuitions raw sense data which the mind processes into genuine experience of external things? If this is so, then intuitions are basically raw sensations, tickles, tinges and feels, and much of what Kant says implies that intuitions must be more than this. Certainly, intuition according to Kant is that element of an experience which is contributed by things outside the mind. For indeed according to Kant experience involves the mind’s bringing the sensory content coming from outside under the various concepts it has at its disposal; it is indeed because he wanted to insist on this that Kant, as Dilthey points out, places such weight “… on his sharp separation … of intuiting and thinking.” (S.149) Experience involves for Kant both the brute impingement of the experienced object on us—our awareness simply of this thing there, given to us now in experience—; and the classification of this thing there under some concept. So seeing the apple tree in the garden consists in 1) our awareness of the tree as some thing so to speak forcing itself upon us as undeniably there, given to us now in experience; and 2) our awareness of the tree as falling under the complex concept or predicate “apple tree in the garden.” If this is right, then intuition for Kant is that part of a full-fledged experience which consists in our apprehension of the object experienced simply as some thing there and now.

Dilthey wants to deny that there is any such sensory content, at least not in the sense in which Kant seems to have conceived it, namely, as a component of experience which is not merely distinguishable, but separable. For at times it does look as if Kant regards intuition as a genuinely separable kind of input which the mind receives from outside and then shapes into a full-fledged experience of something as falling under a specific concept or predicate. Dilthey rejects this idea; he says, “Aber in dem, was er (Kant—B.C.) Anschauung nennt, wirken überall Denkvorgänge oder ihnen äquivalente Akte mit.” (S.149) So in sensory content there is always already, right from the outset, something conceptual and discursive.

In a sense, Dilthey is right to read Kant as maintaining this and he is certainly right to regard the idea of intuition as not merely a logically distinguishable aspect of experience, but also a genuinely separable component, as absurd. I surely cannot apprehend something as a mere this here now; on the face of it I must always have recognised it as falling under a concept. No doubt in some sense I can, on occasions, apprehend something simply as some thing there; for example, I notice it out of the corner of my eye and it disappears too quickly for me to see precisely what it is, i.e., what concept it falls under. But even so, this only seems possible because I do have some concepts under which I could have subsumed it, even if on this occasion I did not have time to. So Dilthey would certainly seem to be right in this much: I can have no intuitions, no apprehensions of things as given to me at points in space and time, without being able to recognise them as falling under some concept. He would also seem to be right in his implicit assumption that at least sometimes Kant does talk about intuition as if it were a separable component of experience which I could have even if I had no conceptual, classificatory abilities at all.

Dilthey goes on to say that rather than distinguishing two separate processes and abilities, one should rather see the distinction between intuiting and thinking as merely different grades, stages or phases in one and the same process. Discursive thinking, that is, our propositionally structured, non-pictorial reasonings about things, is but a higher stage of our numerous, often unconscious processes of consciousness. In a manner which vaguely anticipates things said later by the phenomenological philosophers, e.g., Edmund Husserl (1859-1938), Dilthey says that in our experiences and in our imaginings (which are of course essentially imagined experiences) there are to be found the same elementary processes as in our discursive thinking. These elementary processes include association and reproduction (of ideas, i.e., the kind of thing Locke and Hume spoke of), comparing, distinguishing, noting of degrees, separating and combining, ignoring somethings, focussing on others, and finally the abstraction of general ideas and lessons from particular experiences. (S.149) Dilthey also says that out of these elementary processes our most abstract categories are abstracted, e.g., the notion of an object of experience, of cause and effect, and of substance and accident. (S.149) Kant, says Dilthey, did not need to derive these categories simply from discursive thinking, since they are involved in all thought processes.

The second Kantian distinction attacked by Dilthey is a related one: that between form and content or ‘stuff’ (Stoff). Kant held that the sensory content of experience came to us in basically atomistic, unorganised array; the mind then used its conceptual faculties and abilities to impart unity to this atomistic, disorganised array. In other words, he thought that the role of thinking in the formation of experience was basically to unite the little bits, to give form to the stuff of sensation. Now Dilthey does not appear to deny this picture totally. But he does insist that we cannot conceive this metaphorical talk of form and stuff as implying that there were no internal relations between what forms and what is formed.9 (S.149-150) Once again, this cannot be really understood as a distinction between two separable entities. There must be form in order for there to be stuff and vice versa.

In short, because Kant used psychology in an ad hoc, uncritical way, his unavoidable recourse to psychology was carried by crude and erroneous assumptions. One can only escape such bad psychology by admitting psychology into one’s epistemological work in a clear, conscious and scientifically controlled way. “Man wird die zufälligen Einflüsse irriger Psychologien in der Erkenntnistheorie nur los werden, wenn es gelingt, ihr gültige Sätze über den Zusammenhang des Seelenlebens zur Verfügung zu stellen.” (S.150) Only with a well-thought out descriptive psychology, which clarifies and articulates the precise sense of distinctions like that between intuiting and thinking, form and stuff, can one prevent such chance influences of erroneous psychologising from affecting investigations in the theory of knowledge.

This is an extremely important point: it articulates Dilthey’s conviction that descriptive psychology at least involves some activity of clarifying concepts characteristically used in psychology and epistemology, e.g., experience, in order to prevent false understandings of what these concepts amount to from misleading one. So descriptive psychology as Dilthey understands it has a kind of critical role which exposes mistakes and distortions in one’s grasp of certain things. This idea, namely, that modern philosophy has created certain unnecessary problems for itself by misconceiving in fundamental ways such ‘psychological’ notions as subject, object, experience and the like is well founded: if one looks at, e.g., Descartes, one can quite literally pinpoint in his certain crucial moves he makes which lead to such classically modern problems as the problem of the external world. In his Second Meditation Descartes discusses the nature of sense-experience and establishes, rightly enough, that his experiences of things in the world can be deceptive. On the basis of this, however, Descartes then decides that this means that his experiences are entities or events all of which he could have even if they were all collectively wrong. In other words, Descartes comes to regard his experiences as if they formed their own little world such that they could all exist, he could have his own little world of experience, whether or not the external world presented to him by his experiences was really there or not. Just the same idea underlies the modern thought-experiment of brains in a vat: surely, it is conceivable that a clever computer scientist should connect up a living human brain in a vat of nourishing fluid in such a way that the brain receives exactly the same electrical impulses at its nerve endings as it would if it were the brain of a complete human being getting about their business in the world. Surely, such a brain would have just the same experiences as it would have if it were indeed the person it has been tricked into thinking it is. Here you can see the same assumption as that made by Descartes: that all one’s experiences could be wrong without prejudicing their identity as experiences in any way. Once this assumption is granted, there is no dodging the seeminingly unanswerable question, “How do you know that you are not a brain in a vat?” Descartes envisaged just the same kind of scenario: “How do you know that you are not being massively deceived by a supremely powerful malicious demon into thinking that you are the person you think you are? Perhaps it is all illusion.”

So the descriptive psychology which Dilthey envisages has the character of an attempt to determine the precise nature and structure of our experience in a first-person reflection—‘Selbstbesinnung’ (S.151)—on what it is like for us to have such experience. Such first-personal descriptive analysis is the basis upon which explanative psychology can clarify to itself what it means by the terminology it wields. (S.153) And this role in clarifying basic concepts and terminology extends beyond psychology to all the other sciences which presuppose the psychological, in other words, to all the distinctively human sciences (Geisteswissenschaften). This is why psychology is so important, so basic, for Dilthey. And it can play this foundational role because (what Dilthey calls) the psychical nexus is given to us in immediacy and experienced reality. (S.151) At this point, we can see how Dilthey’s conception of psychology and its significance reaches both back into the past and forward into the future. For, as indicated earlier, the description and articulation of the structures of our experience was an essential element of Kant’s way of doing philosophy. And later both Husserl and his student Heidegger were similarly to regard such description and articulation as the way in which philosophy is most fundamentally (which, it is important to note, is not to say exclusively) done.

The Distinction between Explanative and Descriptive Psychology (Chapter II)

Dilthey points out here an interesting precedent to the distinction he wants to make between an explanative and descriptive psychology. This is the distinction that Christian Wolff (1679-1754) makes between rational and empirical psychology. The details of this distinction are not relevant, for, as Dilthey points out, Kant showed that the idea of rational psychology was dubious. (S.154) What is important is that Dilthey sees a valuable kernel of truth in Wolff’s distinction. Dilthey regards as valuable both Wolff’s distinction between a psychology that describes and a psychology that explains; and his idea that descriptive psychology exercises a controlling function over the explanative kind.

What is particular interesting here is that for Wolff the descriptive kind of psychology is empirical psychology, whereas he associates explanation with rational psychology. This indicates a crucial fact about the origin of the idea of a natural scientifically oriented psychology, a fact which is often overlooked or even forgotten today: this idea stems primarily from the rationalist tradition of early modern philosophy. This is something Dilthey himself alludes to on S.155, where he speaks of the elimination of the metaphysical elements from explanative psychology; he means, of course, the typically rationalist metaphysical notion that psychology studies the empirically unobservable, rational human soul. The roots of explanative psychology in rationalist philosophy may at first seem surprising; surely explanative psychology is the kind of psychology that wants to be empirical, not a priori, like bad old philosophy and metaphysics? But this kind of posture is a decidedly nineteenth and twentieth century attitude which obscures the roots of modern psychology. The conception of nature as a system of an orderly geometric whole, i.e., as governed by strict, mathematically describable causal laws which are uncovered by the method of hypothesis and experiment is much more congenial to rationalism than to classical empiricism. And this picture of the universe as a geometric whole derives explicitly from Plato and Pythagoras, who enjoyed something of a revival in the sixteenth and seventeenth centuries. It is well-known, for example, that Copernicus and Kepler were fascinated by this Platonically inspired vision of the cosmos. So it is wrong to see the rise of explanative psychology simply as the rise of honest, empirically-minded scientists going in to bat for hard, indisputable facts against woolly philosophical speculation. It is in fact just as much the attempt to extend to the psychological, and thereby validate, a decidely metaphysical picture of the universe, namely, as at bottom, in reality, mathematical in character.10

Dilthey says that two valuable aspects of Wolff’s distinction remain, despite Kant’s convincing argument in the Critique of Pure Reason (in particular, in that part of the Transcendental Dialectic which Kant calls the Paralogisms of Pure Reason) that rational psychology as Wolff conceives it is impossible. (S.154) These valuable aspects are 1) the very idea of a distinction between descriptive and an explanative procedures in psychology; and 2) the insight that descriptive psychology is the basic upon which explanative psychology gains access to its kind of empirical data, namely, experience, the psychological nexus, as it presents itself to each of us, and that descriptive psychology exercises a control over the way explanative psychology uses and develops its basic concepts. (S.154)

In the nineteenth century various people expanded on Wolff’s distinction. In particular, Dilthey mentions a disciple of Herbart, namely, Theodor Waitz. According to Waitz, natural scientifically oriented explanative psychology presupposes the results of so-called descriptive psychology. The latter describes, analyses and classifies psychological phenomena as preparation for the subsumption of such phenomenon under general laws formulated in and by explanative psychology. A particular feature of this explanative endeavour to formulate general laws is its psycho-physical character: explanative psychology attempts to establish law-like connections between purely physical phenomena on the one hand and the psychological effects produced ‘in the mind’ by such physical phenomena. The idea is to come up with a model of how the psychological causally depends on the external world, conceived purely physically and thus natural scientifically.

Now while Waitz was prepared to retain the valuable aspects of Wolff’s notion of psychology, he sought, in keeping with the post-Kantian, post-Hegelian anti-metaphysical spirit of the times, to eradicate all traces of metaphysics from explanative psychology. (We have seen how Fechner sought to establish an empiral, natural scientifically oriented psychology which broke with the traditional conception of psychology as the study of the human soul—the Greek for soul is precisely psyché—and thus remained neutral on the issue of the soul’s existence in much the same way as physics is neutral on the issue of God’s existence.) Because of this willingness to break with metaphysical conceptions of psychology, Waitz was, says Dilthey, able to pin down a number of important aspects of the relation between explanative and descriptive psychology. He appreciated the hypothetical character of explanative psychology. He also appreciated the growth in what Dilthey indicates as the various means and techniques of descriptive psychology: here Dilthey speaks of 1) comparative studies, 2) developmental histories of individuals and societies and 3) anthropology.

With regard to 1) and 2): > Ad 1. In and just before Dilthey’s time, there had arisen a tremendous interest in identifying by comparison both the similarities and unique differences between different cultures, languages, religions, histories, folklores and innumerable other things. Wilhelm von Humboldt (1767-1835) investigated the different languages in order to discover traces of the Indo-European mother tongue from which all ancient and modern European languages, except, I think, Finnish and Hungarian, stem and which is apparently the origin of languages like Persian. The Grimm brothers set about compiling accounts of European folklore and fairytales, and also wrote a massive etymological dictionary exploring the origins of German in the Middle Ages and earlier. This desire to identify similarities and uniquenesses, i.e., what made a culture what it as a particular individual was, extended even to comparisons between the social and psychological lives of animals and those of humans.

Ad 2. Tied in with this development of interest in comparative studies was an interest in the way which individuals, societies and cultures developed and formed themselves. Numerous philosophers and other theorists before Dilthey had proposed more or less complex accounts of how individuals and societies develop, e.g., Hegel and Schelling.11 But these were often abstract schemes derived speculatively, that is to say, from the standpoint of philosophical reflection rather than empirically. Hegel, for example, claimed to have established in his Logic a certain principle of maximal self-organisation and coherence which could be found realised in all sorts of phenomena, in particular, in the historical process. For Hegel the task of a philosophical history as opposed to a mere empirical recounting of what, how and why things happened was to take this abstract schema and interpret historical events and processes in its light, so as to exhibit this schema as indeed embodied in the historical process.

Dilthey says (S.156) that there are two respects in which the relation between descriptive and explanative psychology need to be developed further, beyond Waitz. He points out that explanative psychology arose out of the analysis of perception and recollection—see, e.g., Descartes’ Meditations on First Philosophy and Locke’s Essay concerning Human Understanding. From the outset it centred around the idea of sensations, representations, feelings and the like as atom-like elements of psychological processes which combined and interacted with one another in various ways. Dilthey mentions the famous, not to say infamous empiricist notion of association: according to Hume, for example, our minds have a certain tendency to associate certain kinds of ideas with other kinds of ideas. In virtue of this tendency we form concepts of different kinds of identity, e.g., the concept of a dog, a cat, water, and the like, as opposed to concepts of individual properties, like brown, hairy, tasteless, and so on. Others have spoken of how these elements not merely associate, but positively fuse with one another (Verschmelzung). In addition, Leibniz and then Kant spoke of apperception, i.e., consciousness of self as being in this or that psychological state. (Leibniz spoke of such self-consciousness as perception of a perception; this is why he created the neologistic term ‘apperception’. This indicates that apperception is not simply identical with consciousness of self as thus or so; so be aware of self as sitting at one’s computer is not apperception, although being aware of self as having a headache, or indeed as being aware of self as sitting at one’s computer, is.)

Thus, says Dilthey, explanative psychology “… hat … gar nicht die volle Menschennatur und deren inhaltlichen Zusammenhang zum Gegenstand.”12 (S.156) The task must be to bring the complete psychological constitution to the fore. The totality of psychological life, all the nexi (Zusammenhänge) implicated in this psychological life and all its content must be comprehended. As we have already seen, by ‘psychological life’ (Seelenleben) Dilthey does not mean just isolated processes of, say, cognition or perception, but the total psychology of an individual, the overall way this individual perceives, cognises, wills and feels. For Dilthey believes that this total psychological life is not an aggregate of abilities, faculties, skills and psychological states, but an organised totality or whole which is logically prior to and causally formative of any individual ability, faculty, skill or psychological state distinguishable within it. This is of course why he speaks of so often of the psychological ‘Zusammenhang’ or nexus; this nexus just is this logically prior, causally effective whole. Now Dilthey makes an interesting observation in connection with his demand for a descriptive analysis of the psychological nexus which will capture its wholeness: with regard to the volitional and affective side of our psychological life this analysis must somehow account for our striving for preservation and extension of the self, while on the cognitive side it must account for the character of necessity in certain propositions we endorse. Finally, it must account for the character of obligation, of ‘oughtness’ (Sollen) or normative obligatoriness, which characterises certain of our actions. (S.156) According to Dilthey, a purely explanative psychology seeks to explain psychological phenomena like perception, cognition, volition and feeling ‘synthetically’: it identifies the elements which make up the appropriate psychological process and synthesises them into a theoretical model of, e.g., cognition by making hypotheses as to the causal links between, and arrangement of, these elements. He seems to believe, however, that because of this ‘synthetic’ approach explanative psychology cannot explain, or even get into view, certain very fundamental characteristics of the human psyché or self. Firstly, it cannot explain the holistic character of the self’s volitional and affective side, namely, our overall concern for self-preservation and in particular for self-extension. Secondly, it cannot explain the fact that in our cognition of the world we hold certain sentences for necessarily true. I take it that Dilthey does not mean here logical truth or analytical truth, i.e., the kind of truth exemplified by the law of contradiction or the proposition that all bachelors are unmarried males. Rather, he means the truth of such cognitively fundamentally, yet non-definitional, non-analytic propositions as that of causality, i.e., that the world is a causally regular place which works in accordance with the principle “same cause, same effect”. Thirdly, Dilthey thinks that such a psychology will never be able to explain how and why humans hold certain actions to be not prudent or imprudent, but rather morally right or wrong.

It is not at all clear that explanative psychology should not be able even to get into view the phenomenon of an overall tendency to self-preservation when this tendency is understood in a purely biological or evolutionary sense. But I do not think Dilthey understands this tendency in this naturalistic way—as is indeed indicated by his talking in one and the same breath of self-preservation (or perhaps better, self-maintenance) and self-extension. It seems clear enough that by self-extension, i.e., “… Erweiterung unseres Selbst …” (S.156), one could only mean the self-conscious development of characters, abilities, faculties and proclivities which are both novel and above all non-biological. Today, one might call it something like self-realisation or, to use a word both more trendy and more vacuous, ‘flourishing’. It is indeed more than plausible that explanative psychology could not make sense of self-preservation, organisation and extension in this non-naturalistic sense.

On the face of it, it is also unclear why explanative psychology should not be able to explain our belief in the causal order of nature in some biologistic way. There is indeed a school of naturalistic epistemology around which thinks it can provide such an account of these beliefs: so-called evolutionary epistemology. The evolutionary epistemologists concede that our belief in causal order cannot be, as Hume pointed out, an empirically acquired belief like our belief that tomorrow the sun will rise as it always has in the past. They deny, however, that conceding this requires them to regard such beliefs as having no biological explanation—as if the only way to account for them was to take Kant’s route and declare them conditions of the possibility of experience. Rather, they say that such beliefs have been hard-wired into us by evolution; this is why all humans have these beliefs, and this is why we regard them as having some special kind of necessity. As for the truth of such a hard-wired belief like our belief in causal order, well, the very fact that we have this belief shows that it is true. For if it were not true, if the world were not a causally regular place, we would not survived. But obviously we have survived, so by modus tollens this belief must be true. The evolutionary epistemologist says that this kind of account of such fundamental beliefs and their truth is not merely preferable because it is nicely naturalistic, i.e., is part of evolutionary biology and thus constitutes a natural scientific account of the phenomena in question. It is also preferable because it entails that the truth of such fundamental beliefs as that in causal order is not at all necessary. It is just a fact about the universe that it is causally regular; there is thus nothing special about our belief in causal order except the fact that it has been hard-wired into us by Darwinian processes of natural selection. Given that the world is causally regular, it is hardly surprising that intelligent beings should evolve who believe in its causal regularity; they may well not have survived had they not thus evolved.

Does this really show that, pace the likes of Dilthey, there can indeed be an explanative psychological and even biological account of why we believe such things as that the world is a causally regular place? Does it really suffice as an account of why this belief is true? These are actually two questions which need to be distinguished. Perhaps our belief in causality is in some way hard-wired into us for the reasons outlined by the evolutionary epistemologist. This much one can readily concede, for it is an explanation of how a certain phenomenon, namely, our belief in causal regularity, has come about. That is, it is a genetic account of the processes which have led to us having this belief. But does it really give a satisfactory account of the truth of this belief, and thus of why we are justified in holding it.

It is not hard to see that the evolutionary epistemologist account is not satisfactory. It is in fact question-begging. The evolutionary epistemologists argues that the very fact that we have our belief in causal orders shows that this belief is true. For if it were not true, if the world were not causally regular, we would not survived. Since we obviously have survived, it follows by modus tollens from this conditional, that our belief is indeed true. The question is, however, how the evolutionary epistemologist knows that this conditional is true? How does he or she know that if our belief in causal order were not true, then we would not have survived? The naive evolutionary epistemologist will answer, “Well, if there were no reliable natural processes, then we could have made no reliable predictions about events in the world, we could never have grown crops, have hunted for animals, etc., etc., and so our life would have been impossible.” But this is obviously question-begging, for the evolutionary epistemologists is assuming right from the start that we are natural animals, subject to certain lawful causal regularities such as starving to death if we do not eat often enough, thirsting to death if we do not drink often enough, dying if our bodies get ripped apart by lions, etc., etc. So the truth of the evolutionary epistemologist’s crucial conditional in fact presupposes what it is used to show.

If, however, the evolutionary epistemologist ceases to assume that we are natural animals, then the crucial conditional has no justification and the whole argument is invalid. If, e.g., the evolutionary epistemologist assumes that the word ‘we’ denotes us merely in our capacity as beings who believe themselves to have experience of a causally regular world and are now asking why this belief in causal order is true, then he or she can offer no obvious reason why the crucial conditional is true. Kant thought he could provide an exceedingly unobvious reason for the truth of the conditional that if our belief in causal order were not true, then we would not have the kind of experience that we undoubtedly do have. But this is a different conditional and the argument Kant provides for it is decidedly non-naturalistic; it is a piece of so-called transcendental philosophy. This Kantian, transcendental way of justifying our belief in causality is precisely the kind of thing our evolutionary epistemologist claims to have avoided altogether. But in fact the evolutionary epistemological alternative is a spurious one. It only looks plausible because the evolutionary epistemologist indulges in a very classical confusion of early modern philosophy: he or she confuses giving a genetic explanation of why we have certain beliefs with giving a justification of why these beliefs are true, i.e., with giving an account of what right we have to hold them.

It looks, then, that one can certainly find good reasons for endorsing what Dilthey says about the inability of explanative psychology to account for at least two of the fundamental characteristics he associates with the psyché or self.13 Not that the reasons outlined here are ones to be found in this second chapter of the Ideas; to this extent, the account given here of why Dilthey might want to say that explanative psychology cannot account for these fundamental characteristics is a highly reconstructive piece of interpretation. Nonetheless, it does seem to make sense of what Dilthey says about these two characteristics.


  1. Drobisch described Herbart as the Copernicus of psychology and said that we were still waiting for its Newton—see Sachs-Hombach, S.86-87.

  2. Kant, Critique of Pure Reason, B xiv-xv.

  3. See van Gelder, Tim, “What might Cognition be, if not Computation?,” Journal of Philosophy, Vol. 91, No.7, July, 1996, pp.345-381, esp. p.380.

  4. See Aristotle, Nichomacean Ethics, Bk I, Ch.3, 1094b7-28.

  5. See the dedication to Count Paul Yorck von Wartenburg with which Dilthey prefaces his Einleitung in die Geisteswissenschaften, GS I, S.ix.

  6. Of course, one and the same object can be studied by different disciplines: John Howard can be studied by the physicist, for he is a material object. He can also be studied by the physiologist, for he is a living being. And he can studied by the psychologist, political scientist and the historian, all of whom Dilthey regards as engaged in a particular form of human study, i.e., Geisteswissenschaft.

  7. It means “a binding, tying together, entwining, connecting.”

  8. In other words, its influence is hidden and thus what one might call ideological. Descriptive psychology, it seems, has something of the character of ideology critique: its role is to control the hidden influence of one’s tradition and culture by making this potential influence out into the open.

  9. Dilthey uses an example which was to form the topic of much discussion by psychologists and phenomenological philosophers in the late nineteenth century, from Brentano through to Husserl: the perception of temporally extended entities like melodies.

  10. Note that this picture is enjoying a revival these day, in the celebration of non-linear dynamics, or chaos theory, as it is more popularly known, fractal geometry, systems theory and the like. Particularly amazing about this revival is the way it portrays itself as long-haired, warm and cuddly, unlike bad old mechanistic physics of the Newtonian and even Einsteinian kind. Thereby it obscures its fundamental affinity with the older kind of physics.

  11. Characteristic of such thinkers was the tendency already noted to assume that the development of the individual conscious being, i.e., ontogenesis, was structurally identical with the development of the larger society and culture, i.e., phylogenesis.

  12. Dilthey goes on to say, “Daher stellte ich zu einer Zeit, in welcher diese Grenzen der erklärenden Psychologie noch schroffer als heute hervortraten, ihr den Begriff einer Realpsychologie gegenüber.” (S.156) The time Dilthey is speaking of is 1865.

  13. In these notes I was unable to discuss the third fundamental characteristic, namely, the human propensity to regard certain kinds of action as not just wise or unwise, but morally right or wrong. (Added 10.03.16)