SOURCE: "Collingwood and the Idea of Progress," in History and Theory, Vol. XXIX, No. 4, 1990, pp. 21-41.

[In the following essay, van der Dussen examines Collingwood's view of the idea of progress as both epistemological and metaphysical in nature.]

I

At the beginning of the chapter in his autobiography entitled "The Need for a Philosophy of History" Collingwood claims that two branches of philosophical inquiry need special attention. Besides epistemological problems related to historical knowledge he mentions in this connection "metaphysical problems, concerned with the nature of the historian's subject matter: the elucidation of terms like event, process, progress, civilization, and so forth" (Aut, 77). [In the text the following abbreviations are used for Collingwood's works: EPM: An Essay on Philosophical Method; Aut: An Autobiography; EM: An Essay on Metaphysics; NL: The New Leviathan; IN: The Idea of Nature; IH: The Idea of History. From R. G. Collingwood: Essays in the Philosophy of History, ed. W. Debbins (Austin, 1965): CPH: "Croce's Philosophy of History"; SHC: "Oswald Spengler and the Theory of Historical Cycles"; THC: "The Theory of Historical Cycles"; PP: "A Philosophy of Progress."]

Since Collingwood wrote these words, attention has been directed almost exclusively towards the epistemological aspects of history. It has become common to call this branch "critical" or "analytical" philosophy of history, as opposed to its "speculative" counterpart. It is striking, though, that the subjects dealt with by the latter do not correspond to the ones Collingwood mentions under the heading "metaphysical problems." For the usual questions discussed by speculative philosophy of history concern the possible patterns, mechanisms, or purpose of history and not the type of concepts Collingwood cites as examples. One can only wonder why Collingwood's admonition has not been taken to heart and why philosophers of history, on the rare occasions they were dealing with the object of historical study, have confined themselves to the "speculative" subjects mentioned before. For a conceptual analysis of the terms Collingwood mentions—and one could add to these concepts like contingency, unintended consequences, revolution, structure, culture, or nation—would certainly be of interest for a better understanding of the past.

Collingwood himself, however, had a keen interest in these metaphysical questions, and this is especially true of the concepts of process, progress, and civilization. Given the fact that they are also explicitly mentioned by Collingwood as examples of a special branch of the philosophy of history, it is surprising that they have almost completely escaped the notice of the many commentators on Collingwood's philosophy of history. An exception has been L. O. Mink, who has rightly brought to the fore the importance of the concept of process in Collingwood's philosophy of history. Though this doctrine is—in the words of Mink—a "recessive" one in The Idea of History, its significance is illustrated by the fact that Collingwood refers to it in his autobiography in connection with what he calls "my first principle of a philosophy of history" (Aut, 97). This principle was the idea, also expressed in The Idea of History, that the past is not dead, but in some sense living in the present."At the time, I expressed this by saying," Collingwood remarks, "that history is concerned not with 'events' but with 'processes' that 'processes' are things which do not begin and end but turn into one another" (Aut, 97-98). He developed his ideas on this subject at some length in 1920 in the essay "Libellus de Generatione" which "was primarily a study of the nature and implications of process or becoming" (Aut, 99); but unfortunately this essay is unavailable.

In addition to Collingwood's interest in certain "metaphysical problems" there is another reason for stressing their importance, which is that in both the "Preface" and "Introductory Lecture" of his "Outlines of a Philosophy of History" (the "Martouret essay" of 1928), Collingwood emphasizes that history a parte subjecti (epistemological or methodological problems) and history a parte objecti (metaphysical problems) are not only closely related, but are inseparable. This idea is based on the doctrine of the ideality of the past, which is the central theme of the Martouret essay. According to this doctrine the past should not be conceived as an objective reality, but as being intrinsically related to the thought of the historian. It was the error of the traditional "speculative" philosophers of history, in Collingwood's view (though he does not use the term), that they regarded the past as a special kind of reality with special characteristics."But when the methodological view of the philosophy of history is combined with the doctrine of the ideality of history," Collingwood says, "all objection to a metaphysical philosophy of history vanishes. For the necessary forms and conditions of historical thought are now seen to determine the necessary forms and conditions of its object."

It is remarkable indeed to see Collingwood here advocating an idea that only lately has been put forward by philosophers of history such as Haskell Fain and Peter Munz, namely that instead of making the usual sharp distinction between "speculative" and "critical" philosophy of history we should conceive of them as mutually related. In cases like these—and other examples from the unpublished manuscripts could also be given—it is hard to avoid the conclusion that the course of the philosophy of history since the Second World War would have been different had Collingwood's manuscripts been known at an earlier date.

The fact that Collingwood considered history a parte objecti inseparable from history a parte subjecti does not mean that he was of the opinion that the traditional "speculative" philosophies of history should be seen as a valid enterprise. In addition to the fact that they viewed the past as an objective reality, their grand themes suffered, one could say, from violating what Collingwood called "the principle of the limited objective" (NL, 31. 61).

An illustration of Collingwood's conception and treatment of metaphysical problems in the philosophy of history is provided by the course of lectures on the subject which he delivered during the first two terms of 1936. This course consisted of two parts: a "History of the Idea of History" and a "Metaphysical Epilegomena."

It has become famous, of course, because of its posthumous publication by T. M. Knox under the title The Idea of History. It is interesting to notice, here, that while Knox gave the second part the title "Epilegomena," Collingwood titled it "Metaphysical Epilegomena." In addition, Knox in two cases altered the titles of its subdivisions: Collingwood's "Re-enactment of past experience the essence of history" becomes Knox's "History as re-enactment of past experience," and Collingwood's "Progress" becomes Knox's "Progress as created by historical thinking."

The series of lectures under the title "Metaphysical Epilegomena" clearly illustrates how closely in Collingwood's opinion a typical "epistemological" subject like the re-enactment doctrine was intertwined with a "metaphysical" subject such as the idea of progress. (One might suggest that the reason why Knox altered the title of "Progress" and deleted the word "metaphysical" from the "Metaphysical Epilegomena" had to do with his differing conception of the "metaphysical").

II

The importance that Collingwood attached to the idea of progress can be gauged from the fact that he devotes a third of his space to the subject in the epilegomena to his course of lectures. It is also confirmed by the several articles he had already published which deal with the concept of progress: explicitly in "A Philosophy of Progress" (1929), and in some detail in "Oswald Spengler and the Theory of Historical Cycles" (1927), and especially "The Theory of Historical Cycles" (1927), while Croce's views on historical progress are criticized in "Croce's Philosophy of History" (1921).

It is not easy to come to grips with Collingwood's view of the idea of progress. It is less difficult, however, to find the reason for this. For the concept of progress is a typical example of a philosophical concept as elaborated in An Essay on Philosophical Method. In that book Collingwood argues that the character of a philosophical concept is such that, unlike scientific or empirical concepts, no exhaustive definition of it can be provided. A philosophical concept is characterized by an overlap of its specific classes; these overlapping classes each embody the generic essence, but they make up a scale of forms differing from each other both in degree and kind and by opposition and distinction. A philosophical concept is therefore intrinsically unlimited in nature and "leaks or escapes" out of the limits characteristic of non-philosophical concepts (EPM, 35).

The Essay on Philosophical Method can be seen as a treatise in which Collingwood makes clear his ideas concerning the special characteristics of philosophical thought. As such it represents an intermediate stage between Speculum Mentis, published nine years previously, and his later works, the last of which to appear in his lifetime was The New Leviathan, published nine years later.

Though there are similarities as well as differences—both in form and content—between the philosophical studies Collingwood wrote between Speculum Mentis and The New Leviathan, one distinguishing feature remained constant: all of them are informed by a dialectical manner of thought. In An Essay on Philosophical Method the formal elements of this style of thought are worked out in a manner as brilliant as it is clear.

However, there is a problem in that the only explicit attempts made by Collingwood to apply the views expounded in the Essay to particular philosophical problems are to be found in his unpublished manuscripts. Thus, for example, in the "Notes toward a Metaphysic" (1933-1934) Collingwood deals with cosmological problems (especially those concerning the relation between matter, life, and mind), while in his lecture on "Method and Metaphysics" (1935) he deals with the concept of reality. In the published works, on the other hand, Collingwood rarely makes any explicit reference to the philosophical principles laid down in his Essay. These principles, however, are not only embedded in his works, but are also of value in the interpretation of them. This is clearly so in works such as The Idea of Nature or The Idea of History, and of those concepts, that of progress is a fine example.

III

In achieving an understanding of the way in which Collingwood conceives the concept of progress, one has always to bear in mind the way in which he understands other related concepts, for only in this way can the characteristics and specific problems related to the idea of progress be brought to the fore. I am thinking here in particular of the concepts of change, process, development, and evolution. One should further keep in mind the distinction Collingwood makes between nature and (human) history.

The difference between change and development or process (Collingwood in fact does not make a distinction between these terms) is based on the one between matter and life. In a case of change there is always a substratum x which is permanent and changes from one state into another, the cause being something from without. Water might be taken as an example. Though this exists in a solid, liquid, or gaseous state, its essence, Collingwood writes, "is represented by the formula H ² o," the variable in this case being "something extraneous to the generic essence" (EPM, 59).

In a development, however, "there is no substratum and no states of it, but always something turning into something else." This is typical of organic nature, and it was Aristotle who first worked this conception out. Collingwood is not always precise in his terminology, since he says of Aristotle that for him "nature as such is process, growth, change" (IN, 82). After this he continues:

This process is a development, i. e. the changing takes successive forms … in which each is the potentiality of its successor; but it is not what we call "evolution," because for Aristotle the kinds of change and of structure exhibited in the world of nature form an eternal repertory, and the items in the repertory are related logically, not temporally, among themselves. It follows that the change is in the last resort cyclical; circular movement is for him characteristic of the perfectly organic, not as for us of the inorganic. (IN, 82)

The same distinction is also made in The Idea of History, where Collingwood states that two views of natural process are possible: "that events in nature repeat one another specifically, the specific forms remaining constant through the diversity of their individual instances …, or that the specific forms themselves undergo change, new forms coming into existence by modification of the old. The second conception is what is meant by evolution" (IH, 321).

To use the language of An Essay on Philosophical Method, an evolutionary process differs both in degree and kind from a cyclical one. In Collingwood's view the idea of nature as an evolutionary process has been conceived since the end of the eighteenth century on the analogy of the study of human affairs, because by that time historians had begun to see history as a process instead of the succession of separate periods (IN, 9-10).

The theory of evolution itself had, especially following the appearance of Darwin's Origin of Species, an enormous influence on the study of history. As has been pointed out by, among others, Stephen Toulmin, this theory became popular mainly, however, in its vulgarized form as a doctrine of progression rather than as a theory concerning the descent of species as conceived by Darwin himself. Toulmin emphasizes that this progressive or "providential" conception of evolution got its inspiration from Lamarck and Herder and had little to do with Darwin's theory. It is doubtful whether Collingwood was conscious of this fact, since he says of Darwin that he "constantly used language implying teleology in organic nature" (IN, 135). He was, however, extremely critical of the naturalistic progressivism which was so popular during the nineteenth century: "In order to realize the lengths to which this dogma of progress was pushed, it is necessary to go slumming among the most unsavoury relics of third-rate historical work" (IH, 145).

On the other hand, Collingwood took the idea of a cosmic evolution seriously and even saw the concept of evolution as having passed through a biological and a cosmological phase (IN, 133). He was extremely interested in modern cosmological theories, in particular those propounded by Alexander and Whitehead. His extensive "Notes toward a Metaphysic" of 1933-1934—on which The Idea of Nature is partly based—bears witness to this. In these notes Collingwood attempts, very much in the manner of Lloyd Morgan and J. C. Smuts, to develop a theory of "emergent evolution" from matter to life, mind and spirit, with a nisus as driving force.

Collingwood did not succeed in accomplishing a cosmological theory of his own. With hindsight there is no need for us to deplore this fact, nor to endorse Knox's opinion that by comparison with Alexander and Whitehead, Collingwood's "promise never became performance" (IH, xxii); for one important result of Collingwood's endeavors in this field has been his focus on the history of the idea of nature as expressed in the history of natural science. This view is at the present time more in vogue than cosmological theories, though Collingwood's own contribution to it is still hardly recognized.

Collingwood's study of modern cosmological theories had, however, another side effect which was of great importance in the development of his ideas. For though he agreed that nature should be seen as a process, he was unwilling to conclude from this that the distinction between natural and historical processes should be considered as superseded. When considering an historical process Collingwood always refers to the history of mind, that is, to the history Of the human past. In his "Notes toward a Metaphysic" we see Collingwood already developing this view. Although he uses the term "spirit" where he was later to speak of "mind" or "thought," he maintains that:

The life of the spirit is a history: i. e. not a process in which everything comes to be and passes away, but a process in which the past is conserved as an element in the present. The past is not merely a precondition of the present but a condition of it. Whereas in nature the past was necessary in order that the present may now exist (e. g. there must have been an egg that there may now be a hen) the past being thus left behind when the present comes into being, in history, so far as this is real history and not mere time-sequence, the past conserves itself in the present, and the present could not be there unless it did.

In an historical series, Collingwood writes, the earlier continues with "accumulation or enrichment of the existent by the sum of its own past." "For mind in general," he then concludes, "this accumulation is called experience; for consciousness, it is called memory; for a social unity, it is called tradition; for knowledge, it is called history."

So an important result of Collingwood's examination of cosmological problems was the development of a clear distinction between natural and historical processes. This distinction pervades The Idea of History—all manifestations of naturalism or positivism being severely criticized in that book. The argument is represented in its most concise form in the essay "Human Nature and Human History" (IH, 205-231), which was first delivered as a lecture to the British Academy in 1936. In the draft of this essay Collingwood develops the argument point by point, a key part being the contention that not all processes are historical processes, as historical processes are rational and natural processes irrational. Then, implicitly making one of his rare references to the theory developed in An Essay on Philosophical Method, Collingwood remarks that in the relation between the natural and the historical "here, as elsewhere, there is an overlap of classes. Man occupies an ambiguous position. He stands with one foot in nature and one in history." The natural and irrational aspect includes "senses, instincts, impulses and in general the subject-matter of psychology," while the rational aspect includes "the intellect, will and their synonyms and implicates."

From the foregoing it may be concluded that Collingwood's views on history were partly derived from his study of natural processes, and the differences he saw between natural and historical processes. The essence of that difference lies in the fact that nature does not retain its own past in its development, while mind does: nature's past is dead, but mind's past is alive. In the draft of "Human Nature and Human History" Collingwood eloquently expresses this idea thus: "Nature ceases to be what it was in becoming what it is; the phases of its process fall outside one another. Mind, in becoming something new, also continues to be what it was; the stages of its development interpenetrate one another." This view fits in well, of course, with other well-known aspects of Collingwood's theory of history, such as the position that all history is the history of thought, or the reenactment doctrine.

What are the implications of the foregoing for Collingwood's views on the concepts of development and process? Here we can use the theory of philosophical concepts elaborated in An Essay on Philosophical Method as a guide to answering the question. To do so requires that we understand a philosophical concept as constituting a scale of forms of overlapping classes differing both in degree and in kind and being related both by opposition and distinction.

We should, then, place the concept of change at the lowest end of the scale, it being characterized by the idea of a permanent substratum undergoing various changes caused from without. The next stage is the conception of nature as undergoing a cyclical and immanently caused development of both a logical and a spatial kind: from the temporal point of view it shows on the one hand no development, since the forms remain permanent, while on the other hand there is a development from the potential to the actual. Following this comes the concept of natural process or evolution, in which the forms themselves develop. Finally there is the idea of an historical process as the development of mind and rationality—the distinguishing feature of this conception being that in its development mind retains its past in its present.

In An Essay on Philosophical Method Collingwood elucidates the way in which overlap of classes is combined with the scale of forms."Each term in the scale," he writes, "sums up the whole scale to that point" (EPM, 39), for it sees its own point of view as identical with the genus and denies the higher form. The higher form from its side includes the lower one except its denial of the higher form. Applied to the concept of process, this means that historical process as its highest form includes all the lower ones. So it is characterized by being a change, though without a permanent substratum, and of a logical, spatial, and temporal nature, immanently caused and—this being its own increment—retaining the past phases of its own development.

IV

Moving on to the concept of progress, one may observe as a preliminary that it includes the idea of development. It is development, however, of a specific type, which may be described by saying that a later phase is conceived as an improvement upon an earlier phase. As such it cannot be considered as just another increment on the scale of forms of the concept of development, for the concept of progress is also used within the sphere of natural development—although as previously remarked, this is the Lamarckian idea of a progressive evolution rather than the Darwinian theory.

Collingwood emphatically rejects the Lamarckian type of evolution: "The archaeopteryx may in fact have been an ancestor of the bird," he says, "but what entitles us to call the bird an improvement on the archaeopteryx? A bird is not a better archaeopteryx, but something different that has grown out of it. Each is trying to be itself (IH, 322). He was also at times very critical of the idea of progress in history. For example, in 1921 he can be found criticizing Croce for his "vulgar optimism" and for seeing all history as "a change from the good to the better" (CPH, 16-17).

This was not, however, Collingwood's last word on the matter, since he later came back to the question on several occasions, developing rather different views. In this connection it should first of all be noted that he does not always use the concept of progress unambiguously. For though in his essay on the subject in The Idea of History Collingwood indeed rejects the idea of progress in nature in the sense of being a process of improving states, in a preceding passage he maintains that "in one sense, to call a natural process evolutionary is the same thing as calling it progressive" (IH, 321). He means by this that the modifications of the various forms of a natural process can only come into existence in a certain order: "In this sense of the word 'progress, ' progressive only means orderly, that is, exhibiting order." In The Idea of Nature the term "progressive" is used in a similar sense, as for example when Collingwood comments: "where by progress I mean a change always leading to something new, with no necessary implication of betterment" (IN, 14). In his discussion of Kant in The Idea of History Collingwood even plainly states: "All history certainly shows progress, i. e. it is the development of something" (IH, 104).

It is obvious that in these cases the concept of progress is used in a wide sense, equating it with the idea of development as an orderly process. In that sense even natural evolution as conceived by Darwin can be seen as progressive. Collingwood did not dwell on this aspect and preferred to concentrate on the idea of historical process, which has, as we have seen, a rational nature. He considered it an achievement of the first order that in the eighteenth century history was conceived for the first time as making sense: "It had a plot. It revealed itself as something coherent, significant, intelligible," and he described this as "a genuine discovery" (PP, 111).

For Collingwood, seeing history as a plot means that it is conceived as a continuity consisting of the succession of problems confronting human beings and the various solutions found for them."Now such a course of events may be truly called a progress," he maintains, "because it is a going forward; it has direction, everything in it proceeds out of what has gone before and could not have happened without the occurrence of its past" (THC, 86). He then continues: "But though history is in this sense a progress and nothing but a progress, it cannot be so in any other sense. No one of the phases through which it moves is any better, or any worse, than any of the others" (THC, 86-87). The reason given for this position is that each generation is confronted by unique situations giving rise to unique problems.

So we find Collingwood here using the concept of progress in two different ways: one in the sense of history as an orderly and rational process (a "going forward") and one in the sense of each phase of the historical process as an improvement on the last. Progress in the first sense is accepted by Collingwood and even thought necessary, while he rejects progress in the second sense—at least in the passages so far considered. Since progress in the second sense should be seen as its proper meaning, this is the sense on which we shall concentrate.

In order to assess the arguments developed by Collingwood concerning the idea of progress it is important to bear in mind certain relevant distinctions: such as, on the one hand, the distinction between history as an "objective" process and history as conceived by an historian; and on the other, the distinction between history seen "as a whole" and history seen only partially or under a certain aspect. Since Collingwood's essay on the idea of progress in The Idea of History was his latest explicit discussion of the subject, we will consider it first.

In this essay Collingwood discusses, and emphatically rejects, the conception of history as governed by a "law of progress"; for this idea is based, he argues, on the two false and contradictory assumptions of man's superiority to nature and his being subject to a supposed natural law (IH, 322-323). If there is anything like historical progress it should stand on its own merits and not as part of any kind of natural progress.

Progress is a process in which new forms are being developed, and Collingwood describes the idea of historical progress as "the coming into existence not merely of new actions or thoughts or situations belonging to the same specific type, but of new specific types," which should be conceived as improvements (IH, 324). As an example of such an improvement Collingwood asks us to consider a community of fish-eaters which develops a more efficient method of fishing, catching ten instead of five fish on an average day. He is reluctant to call this an "objective" improvement, since the older generation is inclined to consider changes such as these as a form of decadence, while the younger one will see it as progress. The important point here is the reason Collingwood supplies for the impossibility of comparing the two practices within the fisher-community. For the older generation will stick to the old method, thinking it better than the new, and this is not done "out of irrational prejudice," but "because the way of life which it knows and values is built round the old method, which is therefore certain to have social and religious associations that express the intimacy of its connexion with this way of life as a whole" (IH, 325).

The important distinction Collingwood makes here is that between change in respect of a certain activity within the fisher-community (namely catching more fish) and change related to its "way of life as a whole." Though one could claim, of course, that the first type of change can be seen only as an improvement, he is not willing to take this possibility seriously, since it is generally the case that improvements have unforeseen (and often negative) side effects or consequences and that it is precisely this which makes the idea of progress such a difficult one. Here one can only endorse Collingwood's position. A necessary condition of ascribing progress in any particular case is, therefore, that it is conceived as related to the whole of a community's way of life; and in order to do this it is also necessary that both ways of life are known "in the only way in which ways of life can be known: by actual experience, or by the sympathetic insight which may take its place for such a purpose" (IH, 325-326).

In discussing the problem of comparing two ways of life Collingwood deals first with the comparison made by the new generation, and then with a possible comparison made by an historian."Experience shows," he says, "that nothing is harder than for a given generation in a changing society, which is living in a new way of its own, to enter sympathetically into the life of the last," and for this reason "the historical changes in a society's way of life are very rarely conceived as progressive even by the generation that makes them" (IH, 326).

Though one can agree with the first part of this argument, its conclusion is doubtful. For it is certainly not shown by experience that a new generation only rarely conceives its accomplished changes as an improvement. On the contrary, rather the opposite will usually be the case and we will see that at other places Collingwood expresses this opinion also. It is quite another question, of course, whether a new generation has the right to conceive its accomplishments as an improvement. In order to have such a right, Collingwood argues, a person should know his society's past, that is, have historical knowledge of it; for only in this way can the merits of the old and new ways of life be compared."In short," he concludes, "the revolutionary can only regard his revolution as a progress in so far as he is also an historian, genuinely re-enacting in his own historical thought the life he nevertheless rejects" (IH, 326).

Since it may be very difficult for a person of the new generation to have historical knowledge of the old way of life because of his or her lack of sympathy with it, Collingwood goes on to consider the possibility that an historian might function as a neutral judge; and this possibility he flatly rejects. What is required is a comparison of different "ways of life as a whole," and this is not possible: "the task of judging the value of a certain way of life taken in its entirety is an impossible task, because no such thing in its entirety is ever a possible object of historical knowledge" (IH, 327). The reason he gives is twofold, for, he says later, "there must be large tracts of its life for which he has either no data, or no data that he is in a position to interpret" (IH, 329).

However, Collingwood does not mention in this connection a more cogent argument which is relevant with regard to the conception of a way of life as a whole. Even if an historian had no lack of data and was in a position to interpret it faithfully, each way of life should be understood as a unity in which the various aspects are interrelated in a unique way. Though Collingwood is correct when he argues that one cannot speak of progress if certain problems are solved while the solutions to earlier problems are lost, it does not follow that where new problems are solved and the solutions to earlier problems retained we therefore have sufficient reason to assert that there has been progress. In particular, if we are comparing "ways of life as a whole," this being Collingwood's original focus, then the various aspects of a way of life—art, religion, science, technology, economy, and so on—need to be understood in their interrelation, and possible "real" progress in any one of these is a question of a different order.

This is not the only confusing element in Collingwood's discussion of the idea of progress. In the article "The Theory of Historical Cycles" he argues that no phase in history can be considered an improvement on another, for in each phase "men found themselves confronted by a unique situation, which gave rise to a unique problem, or the eternal problem in a unique form" and "to live was to solve that problem, the condition of surviving until the problem changed"; and he continues:

So far as we can see history as a whole, that is how we see it; as a continuous development in which every phase consists of the solution of human problems set by the preceding phase. But that is only an ideal for the historian; that is what he knows history would look like if he could see it as a whole, which he never can. In point of fact, he can only see it in bits; he can only be acquainted with certain periods, and only competent in very small parts of those periods. (THC, 87)

The aspect dealt with here is not the one he takes as his starting-point in The Idea of History; for the point at issue here is whether or not it is possible to conceive "history as a whole," which is a different question from whether a "way of life as a whole" can be conceived: the first concerns the historical process in its entirety conceived diachronically and the second a synchronic conception of a way of life in its entirety.

Collingwood's answer to the first question is clear: we cannot conceive the historical process in its totality, but only parts of it. He deals with this question in his discussion of the idea of historical cycles and argues that the periods with which an historian is well acquainted are seen as luminous and progressive, while the periods with which the historian has little acquaintance are considered dark, primitive, and irrational. This is the background to the cyclical view of history which splits the past up into disconnected episodes of rise and fall, progress and decline (SHC, 74-75; THC, 87-89).

Though this argument is a cogent one, especially in view of Collingwood's conception of the ideality of the past (he does not object to the cyclical view as long as the various periods are not seen as objective realities), it is not related to the question of why a way of life as a whole cannot be grasped by an historian. In fact the latter question—which is crucial to an understanding of the problem of progress—is not answered satisfactorily at all. In The Idea of History it is raised and immediately skirted by using exactly the same argument as that used above (IH, 327-328): but this argument, as we have seen, is related to another question entirely.

Having used an irrelevant argument in addressing the question of whether a way of life as a whole is conceivable, Collingwood nonetheless returns to the question (albeit implicitly) when he develops the view that there is one condition on which the idea of progress "can represent a genuine thought": "The condition is that the person who uses the word should use it in comparing two historical periods or ways of life, both of which he can understand historically, that is, with enough sympathy and insight to reconstruct their experience for himself (IH, 328-329). Here Collingwood explicitly equates understanding an historical period with the understanding of a way of life, the possibility of the latter having previously been in doubt. If for the moment we can forego this problematic element, we find Collingwood using three arguments for the impossibility of comparisons between two different historical periods being made by historians. The first argument is that: "By re-enacting the experience of either in his own mind he has already accepted it as a thing to be judged by its own standards: a form of life having its own problems, to be judged by its success in solving those problems and no others" (IH, 329). The second argument is that it cannot be assumed that "two different ways of life were attempts to do one and the same thing," while the third states that "it would be idle to ask whether any one period of history taken as a whole showed a progress over its predecessor. For the historian can never take any period as a whole" (IH, 329). The reason given for this last argument has already been mentioned (the historian has not enough data or is not in a position to interpret certain data). But this argument is not consistent with what Collingwood maintains elsewhere, since, for example, in his discussion of historical cycles, it is at least not denied, and perhaps even implied, that an historian is capable of understanding an historical period, the latter being equated with a way of life.

If we were to sum up our discussion of Collingwood's treatment of the concept of historical progress, we have to conclude that it shows a lack of consistency, for he denies on the one hand that ways of life can be grasped, while on the other he believes that historical periods may be understood historically. It is nevertheless obvious that Collingwood denies the possibility that historical periods can be compared. The argument he gives is that each period is to be characterized and judged in terms of its own problems and the solutions it finds for them, and that this does not allow comparisons to be made. What is not satisfactory is that he does not properly discuss the basic problem of comparing the unique interrelations between the various aspects within each different period.

v

Collingwood's opinion that historical periods cannot be compared is not his last word on the subject of progress, but his ideas on the subject are varied and not always easy to grasp. It is possible to distinguish four different positions in Collingwood's attitude to the concept of progress: a) It is dependent on a point of view; b) It is meaningless; c) It is meaningful; d) It is necessary.

a) Collingwood's rejection of historical realism implies that any suggestion of historical progress being conceived as an "objective phenomenon" is rejected accordingly. In this connection his usual reaction is to empha-size that not only the idea of progress, but also the idea of decay, is dependent on the point of view taken up by the historian. This emphasis is already to be found in his article of 1921 (long before his elaboration of the principle of the ideality of the past in the Martouret essay of 1928), in which Croce is criticized for his "transcendent attitude" of "asserting the existence of a criterion outside the historian's mind," which implies—at least in Croce's view—that history is seen as a purely progressive process."A change that is really a progress seen from one end," Collingwood retorts, "is no less really a decadence, seen from the other. It is true to say that the decay of archery was the rise of firearms; but it is not less true to say that the rise of firearms was the decay of archery" (CPH, 16).

In "The Theory of Historical Cycles" the same argument is used, Collingwood this time also giving the "growth of the steamship" as an example, it being "the passingaway of that splendid thing, the sailing-ship" (THC, 81). He takes the same position with regard to periods, maintaining, as we have seen, that it is a lack of understanding or knowledge which makes an historian conceive a period as one of decay (THC, 81; IH, 164-165; 327-328) and he speaks in this connection of an "optical illusion" (IH, 328).

Although one cannot compare, in Collingwood's view, periods or aspects of periods in a neutral way, this does not imply that he considers such comparisons to be meaningless. They should be seen as only the expression of a certain point of view, and as long as this is realized Collingwood does not object to them. The cyclical view of history as alternating periods of progress and decay he even considers "a permanent feature of all historical thought." "But wherever it occurs," he adds, "it is incidental to a point of view. The cycle is the historian's field of vision at a given moment" (SHC, 75).

Collingwood does not yield, however, to the "cynical view" that it is only a "matter of taste" "whether you think the course of events is an upward or a downward course," depending "not on it but on you" with the implication "that our preferences are mere matters of chance or caprice" (PP, 109). For he believes that the concept of progress may indeed be used in a meaningful way. Before paying attention to this, however, we should first deal with the cases where Collingwood thinks this concept cannot be used meaningfully.

b) Collingwood is of the opinion that the concept of progress cannot be used in a meaningful way in the realms of art, happiness, and morality.

The reasons Collingwood gives for the impropriety of speaking of progress in art are threefold. The first is that "every phase of art has its own beauty, which it is idle to assess in terms of a scale of degrees" (PP, 110-111). So the various forms of art should apparently be seen as differing only in kind (one may here observe that in An Essay on Philosophical Method a different position is adopted). Using an argument that would return in An Essay on Metaphysics Collingwood maintains that asking whether there is progress or decadence in certain fields of art is a question that "does not arise" (THC, 82; EM, 26); for the various artists were not trying to do the same thing. Therefore, "if we take a single art and study two different phases of its development, we always find them differentiated by a difference of the ideal aimed at" (THC, 82). Of the various ideals it cannot be said that one is inherently better than another, for "a particular age has the task of realising beauty in a particular way" and therefore cannot be assessed "in terms of any other" (PP, 116). In The Idea of History, finally, Collingwood gives another argument for denying that progress in art is conceivable. For a work of art, he here maintains, "is the solution of a fresh problem which arises not out of a previous work of art but out of the artist's unreflective experience" and the flow of that experience "is not an historical process" (IH, 330).

The question whether human happiness has increased or decreased in the past is also considered meaningless (PP, 113). The reason given for this is that happiness cannot be measured and there is therefore no such thing as "the sum of human happiness": "Different ages find happiness in different things" and "the happiness of a peasant is not contained in the happiness of a millionaire," is the simple yet entirely adequate conclusion (PP, 114; IH, 330).

The reason that it is improper to speak of an increase in morality is that "a man's moral worth depends not on his circumstances, but on the way in which he confronts them" (PP, 115). Collingwood therefore particularly objects to the view that certain circumstances such as the abolition of slavery render those living under those circumstances morally more worthy than those who do not (PP, 115-116). The argument developed in this context is similar to the argument used in the discussion of the alleged intellectual superiority of Western civilization as against "primitive" civilizations, for here too Collingwood is of the opinion that "civilized man … mistakes the superiority of his tools for a superiority in himself."

As with art, so in The Idea of History a further argument is given for the denial of the possibility of progress in morality. For the latter is likewise seen as arising out of unreflective experience: "The course of our moral life is conditioned by the succession of our desires; and, though our desires change, they do not change historically" (IH, 330). As we shall later see, however, the possibility of moral progress conceived as expressed in institutions is left open.

So with regard to art, happiness, and morality we find Collingwood explicitly rejecting the idea that the concept of progress is applicable. In this connection it is worth noting that all three are typical of philosophical concepts in the sense developed in An Essay on Philosophical Method. Each of them exemplifies a certain dialectically conceived relation between its classes in the way discussed above. Although they show in this respect a conceptual development, Collingwood denies that there is any historical development, that is, a development in which they could be seen as forming a rational process. This position is a rather odd one, as histories of both art and morality are not only conceivable, but actually written: while one can endorse Collingwood's view that art and morality should be understood as, respectively, the expression of the ideals of a certain age and the confrontation of certain problems of a certain kind, this does not exclude the possibility that man is capable of retaining the experience of these expressions and confrontations from the past. The artistic and moral aspects of the past can therefore be understood as forming a living element of present art and morality.

Whether there can be progress in art or morality—let alone in happiness—is, of course, quite a different question. Though one can only endorse Collingwood's rejection, this does not mean that it is not possible to conceive of a history of them. With regard to art and morality the denial of this possibility is certainly wrong; although it should be added that they should be understood as the history of ideas of art and morality. Moreover, besides being the expression of the ideals of a certain age, art is also a part of that age and of its history; and the same is true of morality and the conditions in which it operates.

Although in The Idea of History Collingwood is of the opinion (as we will see) that progress in science is possible, in the manuscript called "The Function of Metaphysics in Civilization" (1937-1938) he is skeptical of the idea. This manuscript was written in preparation for An Essay on Metaphysics and it is here that the conception of metaphysics as a science of absolute presuppositions is developed for the first time. This conception is indeed relativistic and this is reflected in Collingwood's discussion in the manuscript of the question to what extent our science may be taken as superior to Greek science. He questions this alleged superiority in two ways. In the first place he points out that there is no real standard of comparison between them, though there exists a continuity and development. He suggests that Greek science has provided the spade-work "preparing a soil out of which we moderns are winning our harvests." If this is the case, "is not the richness of these harvests a proof, not of our superiority to the ancients, but of the excellence of their pioneer work?"

In addition, he gives another, more fundamental argument for questioning the superiority of our science over Greek science. Deciding which of two things is better, he argues, implies the possibility of choosing between them; but we cannot be in a position to choose between our science and Greek science, and the question is therefore meaningless. He calls it "a nonsense question," because "to ask it presupposes the existence of a situation which does not exist."

In "The Theory of Historical Cycles" the idea of progress is also linked in a similar way to the possibility of practical choice, for Collingwood there makes the observation that the question whether we might prefer to live in a past period because we think it a better one "cannot arise" as a problem, since "the choice cannot be offered." He therefore says that in speaking of the past "we ought not to call it either better than the present or worse; for we are not called upon to choose it or to reject it, to like it or to dislike it, to approve it or to condemn it, but simply to accept it" (THC, 85). Comparison between historical periods is therefore considered to be both theoretically and practically meaningless.

c) "In its crudest form," Collingwood avers, "the idea of progress would imply that throughout history man has been working at the same problem, and has been solving it better and better" (THC, 84). The identity of a certain problem serves Collingwood as a criterion for the meaningful application of the concept of progress. The absence of such an identity also provides him with his reason for denying the possibility of progress in art or morality, both activities being responses to contingent problems.

There are, however, certain problems which Collingwood does regard as having a continuous historical identity and which therefore allow the possibility of progressive solution. An example is engineering. Discussing the preference one may have for Norman or Gothic buildings, he denies that this can be decided on rational grounds with regard to their aesthetic merits; but if judged by the standards of engineering, the transition from Norman to Gothic was "definitely an improvement": "The main purpose of the architect is to build; the Gothic architect built stronger and cheaper than the Norman" (PP, 110).

Things become rather more problematic when, later in the same article, Collingwood ascribes a similar "objective" improvement to the sphere of political life. Here he speaks of the individual being "progressively liberated… from the tyranny of custom and the crippling weight of a rigid political system" (PP, 119). He is apparently conscious of the possible weakness of this argument, since he goes on to add that "savages would think our political systems and social customs quite as oppressive and inimical to happiness as we think theirs." He escapes from this difficulty by claiming that the social and political institutions developed by us "suit our psychological structure," yielding by this move to a rather fundamental and irrationally based form of relativism. Anyhow, the increase in individual power is reflected in our society, according to Collingwood, in an institutional growth of political power. Both are closely inter-related in his view, and he reaches the conclusion that "The increase in the power of political institutions, which sometimes makes people fear for individual liberty, is thus one of the most certain proofs of human progress, and is both the effect and the cause of an increase in individual liberty itself (PP, 119-120).

Similarly in The Idea of History societal progress is associated with institutional progress. This idea is strongly reminiscent of Hegel, especially as it is linked with the idea of moral progress: "Part of our moral life consists of coping with problems arising not out of our animal nature but out of our social institutions, and these are historical things, which create moral problems only in so far as they are already the expression of moral ideals" and in this sense "there is or may be moral progress" (IH, 330-331).

In the same essay on progress Collingwood maintains that science "is the simplest and most obvious case in which progress exists and is verifiable" (IH, 332). He contends that philosophy and religion may also be seen as progressing, but he makes two reservations: first, any solution to a group of problems should retain the already achieved solutions of past problems. Second, it is only by historical thought that progress can be established: "Whether it has actually occurred, and where and when and in what ways" (IH, 333). This means that progress cannot be postulated a priori, either generally or specifically, as has been done by so many progressivists in the past.

After this point there is a crucial shift in Collingwood's argument, for he now maintains that historical thought should not only establish whether there is any progress, but should create this progress as well: "For progress is not a mere fact to be discovered by historical thinking: it is only through historical thinking that it comes about at all" (IH, 333). The argument he uses to justify this conclusion is consistent with his view of history as a rational process. For we have seen that this process is characterized by retaining its past phases in its present. With regard to science this implies that at a certain phase achievements are kept "by the retention in the mind" and that this is the only way in which progress can take place (IH, 333).

Collingwood's argument is not only convincing, but also fully consistent with his general view of history, as outlined above: both the aspect of continuity and the rational nature of the historical process are emphasized. His position does not, however, appear to be consistent with the theory of absolute presuppositions as expounded in An Essay on Metaphysics. We might put it more bluntly: it flatly contradicts that theory, for according to the theory of absolute presuppositions it is exactly the dis continuity between the various fundamental principles which is most strongly emphasized. These principles are seen by Collingwood as merely succeeding each other throughout history without displaying any rational transition.

It is extremely difficult to conceive of any way in which Collingwood's theory of absolute presuppositions, with its implications of discontinuity, incommensurability, and irrational change, can be reconciled with his essay on progress in The Idea of History, in which he expresses directly contradictory views on the development of science. This contradiction can only be resolved, in my opinion, by keeping in mind the different context within which each argument was developed, and especially the different questions each was intended to answer.

An Essay on Metaphysics (written in 1938-1939) deals with Collingwood's conception of metaphysics. He was extremely concerned by attacks on metaphysics, the latest of which was expounded with great force and clarity by A. J. Ayer, and he was convinced that these attacks were based on misunderstandings concerning the nature of metaphysics. For this reason he decided to make his own contribution to the theory of metaphysics. His theory should not, it is important to note, be considered as an original theory of Collingwood's own making within metaphysics, but rather as a description of what metaphysics in his view had always been. This at least was his expressed intention. Metaphysics is and was, he claims, an historical science, that is, a science describing the absolute presuppositions of a certain time or culture. A metaphysician, therefore, should be a natural observer, who is not in a position to express judgment on the absolute presuppositions he surveys. This implies that any suggestion that any one system of absolute presuppositions is superior to any other is improper, and the possibility of progress in presuppositions is rejected accordingly. As science depends on a system of absolute presuppositions, the possibility of progress in science is also dismissed. As we have seen, Collingwood develops this argument explicitly in the manuscript "The Function of Metaphysics in Civilization," coming to the conclusion that we do not have the right to consider our science better than Greek science.

The essay on progress in The Idea of History (written in 1936 and forming part of the lectures on the philosophy of history) deals with a completely different subject. Here the question is when and in what sense can the concept of progress be applied in a meaningful way. As we have seen, Collingwood is of the opinion that "ways of life as a whole" cannot be compared by an historian for the purposes of evaluation. One could draw a parallel between such "ways of life" and systems of absolute presuppositions in that both appear to be fundamental and all-embracing; and one could go on to conclude that ways of life and systems of absolute presuppositions cannot be judged by an historian, and accordingly cannot be seen as progressive. However, in discussing the possibility of scientific progress in The Idea of History Collingwood does not deal with the assessments made by historians, but those made by scientists themselves. Historical thought is therefore involved, but it is not the historical thought of historians, but that of scientists: "If Einstein makes an advance on Newton, he does it by knowing Newton's thought and retaining it within his own" (IH, 333). Collingwood therefore claims that in order to achieve scientific progress a scientist should be an historian of the subject he studies. Referring to Newton, Collingwood says: "It is only insofar as Einstein knows that theory, as a fact in the history of science, that he can make an advance upon it" (IH, 334).

In this case the past is not viewed, therefore, in the detached way of an historian, but as a participant in the—or better an—historical process. In this way scientists see themselves as participants in the latest stage within the history of science, as historians do within historiography and philosophers within philosophy.

This brings us to the final aspect of the concept of progress—its necessary nature.

d) Not only does Collingwood claim that in relation to certain aspects of the past the historian is justified in employing the concept of progress; he also considers that in relation to solving practical and theoretical problems it is necessary. These problems are always passed down from the past, and in order to be solved they have to be reconstructed and understood by historical thought. In this sense there is a real continuity between the past and the present, but Collingwood refers to it as a continuity "of a peculiar kind" (IH, 333).

What he means is what he refers to in his autobiography as the "incapsulation" of past thought within present thought. This means that present thought is not completely encompassed by the rethought thought of the past, but is conscious of the act of rethinking. In this way a distinction is made between the "primary series" of "real" life and the "secondary" series of the rethought thought of the past (Aut, 113). Collingwood furthermore holds the opinion that all thinking is critical thinking: "the thought which re-enacts past thoughts, therefore, criticizes them in re-enacting them" (IH, 216). It is this critical capacity which allows the possibility of progress, and this is what lies behind Collingwood's remarks when he says of the thought of Newton as re-enacted by Einstein, that it is "re-enacted here and now together with a development of itself that is partly constructive or positive and partly critical or negative" (IH, 334).

The necessary function of historical thought in solving present problems applies not only to theoretical but also to practical problems: having discussed Einstein's advance on Newton, Collingwood observes: "similarly with any other progress," giving the following example:

If we want to abolish capitalism or war, and in doing so not only to destroy them but to bring into existence something better, we must begin by understanding them.… This understanding of the system we set out to supersede is a thing which we must retain throughout the work of superseding it, as a knowledge of the past conditioning our creation of the future. (IH, 334)

As we have already observed, it was Knox who gave the essay on progress in The Idea of History the title "Progress as created by historical thinking." Collingwood refers to this creative aspect only once, in his statement that there is, besides determining whether progress has actually occurred, "one other thing for historical thought to do: namely to create this progress itself (IH, 333). This statement should be interpreted to mean that historical thought is a necessary condition for any form of progress, since it is only from this that scientists, historians or philosophers can see their own work as an advancement. The idea of progress itself could then be understood as having the function of serving as a guiding principle in solving present problems. As such its position is similar to the regulative function of "ideas" in the Kantian sense (Kant's own Idea for a Universal History is a good example of this use). This is made clear by the way Collingwood concludes his article on "A Philosophy of Progress":

The question whether, on the whole, history shows a progress can be answered, as we now see, by asking another question. Have you the courage of your convictions? If you have, if you regard the things which you are doing as things worth doing, then the course of history which has led to the doing of them is justified by its results, and its movement is a movement forward. (PP, 120)

The conception of progress as a necessary idea in guiding our actions is also to be found in Collingwood's discussion of the concept of civilization, for civilization is described in The New Leviathan as "a process of approximation to an ideal state" (NL, 34. 5). And in his preparatory manuscript to that book he goes so far as to say that "civilization and the advancement of civilization are one and the same. The will to be civilized is identical with the will to become more civilized." In his Martouret essay Collingwood states similarly, though more explicitly, that "progress is universal because ideals are always progressively realised. A people which fails to realise a certain ideal is a people which does not regard that as an ideal."

That the concept of progress may serve as a guiding idea is also exemplified in Collingwood's work itself, when he closes his Essay on Philosophical Method by expressing the hope that through the methodological principles enunciated in that book "philosophy may … set its feet once more on the path of progress" (EPM, 226).

When the idea of progress plays such a crucial role in solving present theoretical and practical problems, it is also the task of the historian to reconstruct the idea as it was conceived in the past. Having stated in the passage quoted above from the Martouret essay that progress is to be seen as a necessary aspect of history a parte objecti, Collingwood therefore also goes on to express the same principle a parte subjecti, that is, as an historiographical principle."Progress, then," he says, "is universal in the sense that a narrative of any particular historical period, as it proceeds, reveals more and more clearly the nature of that period's ideals; and it is by these ideals that it ought to be judged."

The concept of progress is an illustration, therefore, of Collingwood's principle that history a parte objecti and history a parte subjecti cannot be separated: the metaphysical questions and the epistemological questions are joined in a necessary unity. But this is not all: it has become clear that many of Collingwood's concerns are united in this principle—for example, the question of the relation between theory and practice. In his discussion of the idea of progress Collingwood raises many important issues and displays many brilliant insights, and for this reason alone is well worth our continuing attention.