By Paul Newall
History may not seem to have much to do with philosophy but—just as we have already seen with science, politics and art—it relies on philosophical assumptions and concepts as much as any other subject. In this discussion we’ll introduce some of the philosophical issues within history and hence try to gain a deeper appreciation of it. First, however, we need to know what we’re dealing with.
What is History?
This may seem like a straightforward question but often an equivocation is made between two distinct uses of the word:
History as the past; and
History as an account of the past.
These are quite different. The first is what we mean when we say “it’s all history now”, which becomes obvious if we just rephrase it as “it’s all in the past now”. The second, on the other hand, is implied when we talk of the history of the Great War, say, or the history of science. This distinction is sometimes quite subtle: when we refer to the history of a period or event we mean not just what happened (the past) but also how and why. Some thinkers have suggested that a way to clear this up definitively is to use history for the second meaning and simply call the past the past.
What is history, then? In the first instance, the past would seem to be just the past: what happened before, whether in a specific period or just generally before now. (An interesting related question is to ask whether the past exists or not.) The problem arises when we try to decide what history is in the second sense. According to the historian Elton:
The study of history … amounts to a search for the truth.
As a consequence of this perspective, we could say that history is the true account of the past. We have already seen that there are different understandings of truth, but in this case we are speaking of a correspondence between what actually happened in the past and an account of it. Later we will look at whether this conception of history stands up to scrutiny and, if not, what could replace it.
Another question we could ask is “what is the purpose of history?” That is, what is it for? Why do we study history in the first place? There are several possible responses:
For its own sake;
To find out the truth about the past;
To try to understand where we came from;
To try to understand why a particular event happened;
To find historical laws;
To justify actions in the present.
We will consider difficulties with some of these below.
What is the Philosophy of History?
The philosophy of history is concerned with the concepts, methods and theories used in history; on the other hand, historiography is the study of the writing of history. When we analyse these we can begin to say something about what history is, as well as what it is not or cannot be. A distinction is generally made between two branches of the philosophy of history: speculative and critical. The latter is concerned with investigating those things already mentioned, while the former tries to find a pattern behind historical events—hidden from sight, as it were, until the historian discovers it.
To appreciate where the philosophy of history differs from and expands on history itself we can refer to Hayden White’s explanation:
The principal difference between history and philosophy of history is that the latter brings the conceptual apparatus by which the facts are ordered in the discourse to the surface of the text, while history proper (as it is called) buries it in the interior of the narrative, where it serves as a hidden or implicit shaping device…
Although this may seem confusing, the important part is the emphasis on “conceptual apparatus”: according to White, the philosophy of history brings to light the implicit assumptions that historians rely on and that – more importantly, perhaps – have consequences for their accounts. We shall examine some of these now.
If we go into the history section of a good bookshop and look around, we tend to find plenty of titles on the same familiar subjects: wars, revolutions or other so-called defining moments. In a large or particularly high quality store we can see that there are histories of all sorts of things and all kinds of people (although we search in vain for a copy of the much sought after academic volume Funny Things Hugo Said). However, we do not see all of history: people, places, events and periods are left out—as they must be, given that there are only so many historians, so much time and so many records to look to. This to say that history is always less than the past. After all, who is writing the history of what we are doing right now?
How do we decide which histories are written, then? Obviously there are commercial considerations to bear in mind, but the academic papers that tend to be the basis for the more popular accounts are not so constrained. How do historians choose what to write about (and how to do it – historiography), apart from the straightforward criterion of something that interests them? For some historians this is an easy question: they work on significant issues from the past. Why the French Revolutionaries decided to act is significant, while what they ate for breakfast is probably not.
An objection raised in recent times, especially by so-called postmodernists, is to ask who decides what is significant: who or what is worth the historian’s attention? Although the example above may seem trivial, they say, not everything is so clear-cut and the allocation of significance is a value judgement. In particular, some groups are very much underrepresented—such as women and minorities. Indeed, given the sheer number of women who have lived in the past, it is hard to argue with feminist claims that women have been excluded from history in almost systematic fashion.
Already, then, we can see that some of the high aspirations for history may not be so easy to maintain. Nevertheless, there is another issue that follows immediately: how do we address this imbalance in history, deliberate or otherwise? Feminist historians, for example, are trying to reappraise the role of women in the past; but this means that they are writing with a purpose in mind. Some philosophers of history suggest that this is not limited to marginalised perspectives but that ideological positions are inevitable. Later we’ll consider some of the arguments for why this is so, but for the time being we can note that it would imply that our original “what is history?” becomes “what is the aim of a particular history?”
Explanation and Description
Another distinction made in the philosophy of history is between history as description and history as explanation. Those advocating the former suggest that the role of history is only to describe what happened in the past – this much and no further. Others say that history does (or must) do more: it must go beyond description and explain why an event happened as it did (or at all). Thus an account of what occurred in (and before) the French Revolution is not enough—it also has to explain why the Revolution happened at all, not least because there appears to be no contradiction or impossibility in supposing that it might not have.
According to some such thinkers, history as description is like bookkeeping; but someone else has to come along and check the figures to see what the sales mean and to understand why people bought one thing and not another. Although the entries (or “what happened”) are vital, they are not enough to be history.
If we take it as given that the historian has to provide an explanation for an historical event, does it make sense to talk about historical causes? As we saw in our thirteenth discussion, causation is a difficult concept with many associated philosophical problems. Even so, one place we can start is to distinguish between necessary and sufficient causes via the more general notion of necessary and sufficient conditions.
A necessary condition is one that must be satisfied before we can say that something belongs to a class. Much like a guessing game, then, if someone is thinking of an animal that happens to be a horse, we could ask lots of questions that give us the conditions that are necessary for something to be a horse. For instance, a horse has:
… and so on. If an animal is to be a horse, these conditions must be satisfied. An animal without hooves cannot be a horse (unless some notorious wit is thinking of a seahorse). A question like “does it have a mane?” answered in the negative would tell us that the animal cannot be a horse (or a male lion, and so on) because a necessary condition for being a horse is having a mane.
A sufficient condition, on the other hand, is one that is enough to conclude immediately that we have—in this example—a horse. If someone asks, say, “does the animal compete with rider in show jumping?” and receives an answer in the affirmative, we know itmust be a horse without any need for further questions. Thus this answer suffices to conclude that we have a horse.
This is a simplistic instance because we do not say that a horse with only three legs is no longer equine. In general, a necessary condition for x to be a y is one of potentially very many that have to be satisfied before we can say “x is a y”, while a sufficient condition is one that includes all the necessary conditions and is enough on its own. To return to historical causes, how far back do we need to go and how wide do we need to look before we can speak of what causedan event to happen? Suppose we take an example like the advent of science and ask, “what caused the rise of science?” Historians of science say that this is a vague question, but necessary causes would take the form of a list of things that were, in the judgement of the historian, required before science could develop. A sufficient cause, however, would be a single event that could bring about science on its own. Almost immediately we can see that the latter course is too ambitious: historical events, it would seem, arecomplex; that is, they are the result of many different factors, so that to look for just one as a cause is perhaps a mistake (although we might speak of more or less important factors).
Nevertheless, another problem with historical causes is that the notion of causality has been brought into history from science and some philosophers of history feel that this was a mistake. The main difference, they say (apart from the epistemological problems we will come to later), is that the actions, motives and other foibles of people are involved in historical events, unlike causal chains in science. When we say that an illness was caused by a virus, for instance, we mean that there was a link between the two that did not depend on the political opinions or upbringing of the person getting sick, say. If, on the other hand, we want to say that the French Revolution was caused by Royal excess, it doesn’t explain much. Why did Louis XIV act in one way and not another? What was the influence of his childhood, or his advisors? What of all the other people involved? And so on. The causal chain is rendered far more complex by the involvement of the human factor, or so the argument goes.
Since history (or, more accurately, the past) is continuous, when can we stop and say that a cause has been found? The difficulty lies in ending the quest for causes in a way that is not arbitrary or according to the whim of the historian. One response is to suggest that we have a cause (or set of causes) when we have enough to offer an explanation of an event. The philosopher of history R.G. Collingwood proposed that a necessary cause in historical investigation is one such that without it the subsequent actions would make no sense. Similarly, a sufficient cause is one that would make the course of events that followed considered “rationally required”. That means, for example, that a necessary cause of the Boer War would be one any explanation of the war must include to be convincing; while a sufficient cause would be one that, once it happened, would seem to make the war inevitable.
Expanding on the question of historical causes and continuing the parallels with science, some historians and philosophers of history have claimed that it is possible to find historical laws, meaning much the same as we do when we talk of scientific laws. An historical law might take the form “whenever x happens, y is bound to follow”; so that, for instance, it could be claimed that “states always turn to war when their resources are insufficient for their population” is an historical law. For those who suppose that it is meaningful to talk of such laws, historical investigation would be the way to check the claim.
Several objections have been made to the very idea of historical laws, of which Popper’s The Poverty of Historicism is perhaps the most famous (historicism being, in this case, the belief that historical laws exist). We have already seen that some philosophers find laws to be problematic. Another complaint is to say, with Oakeshott, that history is always concerned with the particular, not the general. In reply, it is said that occurrences in science are no less unique; but what is sought is the general case that can be described with general concepts. Since history uses these just as science does—with terms like “revolution”, “conflict”, and so on—there is no reason to suppose that the search for laws must fail.
A further criticism is to say—again—that history is concerned with the actions of people and that hence an historical law would have to account for the reasons why a person acted as they did. In response it is said that laws have the form “a person, acting in a rational way in situation A, will invariably do B”. In this way A and B constitute the reasons for acting and the action itself. This is not to say that an irrational person may not do otherwise or that other reasons may change the situation, but only to generalise empirically.
Karl Popper took a distinct line of attack. The error in supposing historical laws to exist, he suggested, lies in supposing history to be similar to science when it differs in one crucial respect: scientific laws apply to closed systems, whereas history—composed of the actions of individuals—is neither closed nor even a system at all. Moreover, the growth of scientific knowledge added to this point: since knowledge has an effect on human behaviour and hence history, we can only predict history via laws if we can also predict the growth of knowledge. If we could do that, however, we would already know it. As a result, there can be no historical laws.
Facts in History
Given the importance of “what really happened” to history, it makes sense to ask if matters are as clear-cut as perhaps some people (including historians) suppose. Here we’ll look at the uses that facts in history are put to and if we can say that there are such facts in the first place.
Facts and Interpretation
It seems a commonplace that we have historical facts to work with, such as “there was a world war between 1939 and 1945”. Even so, these apparently simple facts are not the business of history; instead, it is their combination as explanations that we have seen is taken (usually) to be the historian’s task. However, a question asked by philosophers of history is how much of history is fact and how much interpretation? Since facts themselves are silent, goes the argument, the historian must interpret them to understand their meaning. This interpretive dimension is unavoidable and is added by the historian—it is not “already there”, like the facts are supposed to be. This suggests that we can never get past interpretation to the ultimate meaning or definitive account of the past.
Generally speaking, working historians tend to be unaware of this concern or remain unconvinced by its import. Although interpretation goes on, they say, most facts are not disputed or subject to contention and there is wide agreement about the majority of historical issues. When debate takes place amongst historians, it is at the margins—around a central core agreed by (almost) everyone. For example, most of the facts about the Second World War are known, with discussion not really calling much of this body of knowledge into doubt.
The difficulty with this response is that it overlooks a glaring assumption: namely, that this centre is fixed. Instead, it lies on a spectrum of possible interpretations of the same facts. An example given by Jenkins is that of historical accounts in the old Soviet Union, in which the facts about the Second World War were interpreted from an agreed centre that differed significantly from the centre used by Western European historians. The mistake lies in supposing that a particular centre is the only possibility. The problem of interpretation comes up again on another level when we ask how one centre comes to dominate historical discourse, rather than another.
A difficulty of an altogether different order arises when we begin to look closely at historical facts. To begin with, the term “facts” is loaded: what historians are actually confronted with are fragmentary accounts or traces of the past that are subsequently organised into facts. As we saw in our sixth discussion, facts are theory-laden; and for historians they are doubly so, as it were. The historian constructs an account of the past from other accounts, the evidence he or she refers to consisting in the accounts left by others. These accounts record not facts but what people in the past considered important, selected, interpreted and given from their particular perspective.
We will dwell on this area because of its importance. Consider:
The records we have of the past are incomplete and must always be so.
People in the past did not record everything, any more than we do today.
The historian relies on the observation and memories of others in the past for the accuracy of these records.
The past has gone and hence cannot be recalled to check the accuracy of our accounts of it.
The past is studied from a modern view, using contemporary concepts and understandings.
Several of these are specific concerns that we will return to later.
The problem for the historian is that there is no way around this epistemological issue. If he or she tries to check the truth of an account by it correspondence with “what actually happened”, this appeal is found to be empty. Unlike science, where reference is made to reality, there is no historical reality within reach: all we have are traces of the past, accounts of others that may or may not be accurate. In the absence of any way to say whether they are or not, can it be meaningful to speak of historical truth? We will come to this question below, but for now we can note that the only way to check an historical account is by comparison with others. Thus the historian is forced, as it were, into retreating to a coherence theory of truth. The traces we have can function as limits to interpretation, such than any history has to take them into account (whether by incorporating them or discounting them, with reasons for both), but they cannot determine which of a multiplicity of possible histories within the boundary provided is more accurate. In a sense, then, we have the problem of under-determination from the philosophy of science that we studied before, only much worse.
Language in History
These philosophical concerns may be all very well, but do they really impact on history in a significant way? One way to see that they do is to look at the language used in historical accounts and ask if it possible to use a neutral, value- (or theory) free language to discuss the past. The answer, perhaps unsurprisingly, is no: the words we use reveal perspectives because of the epistemological problems identified above.
A well-known example is the adage that “one man’s terrorist is another man’s freedom fighter”. Should an historian call the crossing of an army from one state to another in the past a war, a disagreement, a liberation, or any number of other possibilities, none of which are theoretically neutral? Is an internal conflict an uprising, an insurrection or a revolution? Is calling it a conflict already to prejudge it? Even something as apparently straightforward as a World War is only obvious to those that share the interpretive framework and may not have the same meaning for everyone—Bushmen, for instance. We can say that the historian describes the event in a way enjoined upon him or her by the evidence, but—as we said before—the records from the past are silent and do not insist on any particular reading. Moreover, the same problem was present for those who recorded events in the first place.
The historian can try to tread a fine line, attempting to avoid describing events from the past in loaded terms, but the very act of composing an account reveals choices made. Consider, for instance, an art historian: by deciding to give the history of a painting, he or she presupposes implicitly that the work is art—not trash. We have seen in our seventh piece, however, that deciding what is or is not art is far from simple. As soon as the historian opens his or her account, decisions are made about what to include or exclude. This leads us, then, to the question of historical method.
According to Hayden White:
… the so-called ‘historical method’ consists of little more than the injunction to ‘get the story straight’ (without any notion of what the relation of ‘story’ to ‘fact’ might be) and to avoid both conceptual overdetermination and imaginative excess (i.e., enthusiasm’) at any price.
In this section we’ll look at the situation within history and see if it is as bad as White insisted.
When we look for the historian’s method we are faced with the same problem as the similar quest for the scientific method: an overabundance of choices. Jenkins makes this painfully clear when he asks:
… would you like to follow Hegel or Marx or Dilthey or Weber or Popper or Hempel or Aron or Collingwood or Dray or Oakeshott or Danto or Gallie or Walsh or Atkinson or Leff or Hexter? Would you care to go along with modern empiricists, feminists, the Annales School, neo-Marxists, new-stylists, econometricians, structuralists or post-structuralists, or even Markwick… to name but twenty-five possibilities?
Each of these (and more besides) is an example of a methodology that is consistent, gets results and is profitable for its users. Unfortunately, however, the epistemological difficulties identified above make a choice between them a tricky matter: what criteria should we use to decide which, if any, is the “best” method? We cannot compare their accuracy in getting at the past because there is no such beast.
Unlike science, then, where we can at least try to say that experiment is better than guesswork by reference to something like reality, with history we have nothing to appeal to but other accounts. We might propose that the structuralists explain something better than the feminists, say, but that can only mean that the explanation accords with most or all of the available records of the relevant past and that the account “makes sense”, explaining matters satisfactorily. None of these terms (“accords with”, “makes sense” or “satisfactorily”) can be given a rigorous definition precisely because a history can only convince subjectively within the boundary set by the traces of the past we have. It can never go beyond them and invite comparison with “what actually happened.”
In summary, there are historical methods but no historical method. The same goes for science and hence this should probably not be surprising, reflecting the breadth of history rather than a shortcoming.
Sometimes we hear the complaint that an historian is not ideologically neutral. What we can learn from the discussion of method, however, is that there is no neutral position from which to do history. It may be the case that an historian distorts (or outright lies about) his or her sources, thus going beyond the boundary set on his or her account by the records of the past, but otherwise history from one perspective is no closer to the past than from another. The complaint that a particular history is based on ideology is rather hollow, then.
Perhaps a less ambitious understanding of the role of ideology in history is to note that people—not just historians—use history as a means to ground or legitimate themselves? Where we have come from can tell us where we are going or justify claims we want to make in the present. We see this practice often enough in attempts to validate the assertion that a country (or crown) justly belongs to one group and not another, or even in the popularity of family trees.
We might want to call a Marxist history of Europe ideological, but why are the alternatives any different? Each seeks to understand the past from within an inevitable framework. As we touched on above, the choice of one word (“invasion”, say) instead of another (“liberation”) only makes sense within a perspective that leads us to choose one and not the other. Rather than dismiss certain ideologies, then, perhaps it would be better to examine them and hence try to counteract the unavoidable influence of our own?
The historian has a potential way out of these concerns, however: empathy. By studying his or her sources in great depth and at length, it is said, the historian can begin to empathise with his or her subject(s) and gain an understanding from their perspective. This is the historical skill or tool that helps avoid many of the epistemological and other difficulties and grants the historian a privileged ability to say what motivated people in the past and why they acted as they did.
There are several reasons why philosophers of history find this wholly unconvincing. The first is the general philosophical problem of other minds, in which it is asked how we can ever know the content of another mind; that is, what someone else is (or was) thinking. This is compounded by the distance between the past and the historian. Another objection is revealed by Croce’s dictum that “all history is contemporary history”, which is to say that although historical sources are from the past they must nevertheless be read in the present. This makes the historian a translator of meaning, but he or she has to do so from his or her own perspective that—as we have seen—is never neutral. In like fashion, Dewey wrote that “all history is necessarily written from the standpoint of the present”. Given that the historian is using contemporary concepts, methodologies, epistemological assumptions, modern understandings of words, and so on, how can these be fully (or partially) shed to empathise with those in the past?
A charge often made against historical accounts in criticism is that they are guilty of anachronism. Perhaps the best way to appreciate what this means is to use an example.
Some historians of science point to the work of Newton and note that, in addition to his work on mechanics, mathematics and other areas for which he is famous, he also spent the better part of his time studying alchemy and biblical prophecy. According to some, this is at best a shame and at worst a tragedy: imagine what Newton could have achieved if he had not wasted his time on the latter subjects, putting all his efforts into the former.
The problem here is that contemporary ideas or values are projected backwards: although we may think that alchemy is a hopeless endeavour (or we may not), that is not to say that Newton did. A similar question asked in his time (“think you alchemy a waste of time, sir?”) may or may not have been answered differently, but since we do not know what he thought (except insofar as we could guess that his efforts suggest he would not agree) we cannot say that he should have acted otherwise without being anachronistic.
From the discussion of empathy we can see that a certain amount of anachronism is unavoidable. Nevertheless, the value judgement that alchemy is worthless is not forced upon the historians by the records he or she has of the past, hence the objection that to say so is anachronistic.
Truth in History
At this point in our discussion, the notion of truth in history seems to have taken a battering. Now we’ll look at possible ways to save it and see if we can breathe life back into it.
Truth as a goal
Earlier we learned that some historians consider their task to be the search for the truth. In spite of the apparent impossibility of ever achieving that, they still maintain that it is worth aiming for. However, if—as we have seen—the truth is not a meaningful concept in history, how can striving for it fare any better?
Thinking back to our long look at truth in our tenth piece, what we see is that these historians are employing a correspondence theory—trying to match up the past and our accounts of it. Whatever we think of correspondence (or semantic) theories in general, it is at least clear that they are inappropriate for history. Instead, the realisation that the only way to test historical accounts is by comparison with others suggests that history requires a coherence theory, with Joyce, Appleby and Hunt calling for “well-documented and coherently argued interpretations that link internally generated meanings to external behaviour”.
Given that the historian is faced with nothing but traces of the past, combined and recombined into accounts but never any more than that, he or she can try to construct a new account that coheres with what is available. As further sources are found, the process begins anew and some previous accounts may be shown to be false. As we found when discussing truth, this gets the historian no closer to “what actually happened”, but what it does do is follow the way he or she works with the available material.
Critics of this understanding suggest that the historian is actually working with a pragmatic theory of truth. History is linked, like truth, to power, with accounts serving to support or undermine dominant or marginalised histories. On this view, truth and falsity serve to shut down interpretations that do not accord with what is useful for a society or group.
Another important concept in history is bias, the idea that traces of the past or accounts of it can be intentionally distorted to serve the purposes of the historian. However, bias only makes sense alongside the similar existence of unbiased accounts; that is, with the assumption that true stories exist that correspond to the past and from which biased versions differ. Since this has been thoroughly undermined, there being no neutral position from which to judge the degree of difference, where does it leave bias?
In some sense, as we said, we can identify where an historian has gone beyond the limits of interpretation given by his or her sources. However, histories that do not rely on a correspondence theory of truth can speak of failing to cohere with other accounts or say that using history in different ways need not be biased but just a difference in goals or methods. In general, if the problem of bias is present within all histories then—again—perhaps a diversity of approaches can help appreciate what historians can achieve instead of striving after correspondence?
Philosophies of History
In our final section we come to speculative philosophies of history—attempts to find patterns in or a structure to history. We’ll consider two general approaches to take to history and then look at two classes of theory in the philosophy of history.
The notion of historical realism is analogous to its scientific counterpart and supposes that the concepts and theories employed in history get at reality—in this case, historical reality or “what really happened”. In particular, the past exists independently of what we think of it. It relies, as we might expect, on a correspondence understanding of truth: even if a particular theory (or account) may not be true, it is more or less accurate by comparison and the aim of historians is (or should be) the truth.
As we have seen above, and as a survey of the scholarly literature within historiography would show, historical realism is a thoroughly discredited position, often disparaged as naïve realism (in the pejorative sense). Nevertheless, there are still very many historians who adopt it and some philosophers of history have lambasted their unwillingness to face up to the failings of realism. However, still others advocate a much-reduced conception of the kind of objectivity that is possible (“defined anew as a commitment to honest investigation, open processes of research, and engaged public discussions of the meaning of historical facts” for Joyce, Appleby and Hunt) and point out that few practising historians today ever believed in this kind of realism in the first place.
In opposition to the realists, having accepted the criticisms given, historical anti-representationalists contend that the correspondence theory of truth within history has to be given up and the constructs of historians understood as fictions, not closer and closer approximations of the past as it happened. They may suggest that a coherence theory of truth is more appropriate or that talk of truth should be dropped completely, “what actually happened” being ultimately meaningless within history since it is forever inaccessible. Historians’ accounts are to be read as attempts to organise the available traces of the past in a coherent way, not to latch on to something that cannot be found.
Much work is still to be done in responding to anti-representationalist ideas, particularly with questions relating to the ancient world. Anti-representationalists hope that a history that can come to terms with its limitations will provide us with more interesting and significant accounts of the past.
Some philosophers of history, most notably Hegel, have proposed that history proceeds in a line—hence linear—and so is directional, or “going somewhere”. For those holding to a linear theory, history is a process that unfolds towards a final goal. This is a progressive view in which what came before was in a sense more “primitive” than now, while what will follow will be an advancement, until such time as the limit is reached. A quote from Hegel that gives a nice example is his remark that:
… the Eastern nations knew only that one is free; the Greek and Roman world only that some are free; while we know that all men absolutely (man as man) are free.
On this view, then, the development of the notion and application of freedom is an instance of a linear advancement.
Although the concept of teleology (discussed in our fifteenth piece) has come in for much criticism when applied to life, many people do seem to feel that we can justifiably say that we have progressed from the past and, moreover, that this is likely to continue into the future. For linear theories this is an inevitability—the playing out of historical laws or plans—which is separate from the idea that progress is contingent: it has occurred but need not have. A further distinction is to ask whether we should say that progress is strictly linear or whether a civilisation (or history in general) can advance and regress, showing a pattern of progress overall but not necessarily in all specific periods. The objections made to historical laws also apply to any speculative philosophy of history.
Another class of theories holds that history proceeds in cycles. The philosopher of history most commonly associated with cyclical theories is Toynbee, who suggested that all civilisations showed a similar pattern of growth, dominance and decay. Using examples from ancient history, he divided the past into several complete civilisations and tried to demonstrate that they each arose through responding to challenging circumstances, developed into fully-fledged societies before eventually crumbling. He used these case studies to look for patterns and hence derive historical laws.
In criticising his work (which, at ten volumes, is far too extensive to effectively summarise here), it was pointed out that it is unreasonable to suppose that general laws could be found on the basis of at most thirty-two examples. Another, more significant problem is that civilisations—not clearly defined by Toynbee—do not exist in isolation and continuation between them is not accounted for in positing their demise. Perhaps the most damning aspect to his work, however, was his refusal to announce the doom of our own civilisation when his studies—if we accept their conclusions—pointed to that conclusion with no likelihood of reprieve.