philosophy - history - epistemology

"But it seems that since either Hitler or Stalin was bound to win one of them was bound to be validated."

“It is difficult to learn what constitutes a philosopher because it cannot be taught. One must ‘know’ it from one’s own experience, or else have the pride not to know it. But all the world today talks about things which it cannot have experienced. And this is most true, alas, of philosophers and all the circumstances surrounding philosophy.”

In London, street crossings have a painted warning for pedestrians to look right for traffic is coming from the left lane. Since left-side driving in the UK is the law, this indication would seem to be redundant. Any fool can quickly deduce towards where he should look when getting from one side to the other side of a London street. But the fact remains that in London, traffic authorities apparently consider people fools. The thought comes to mind that, since the vast majority of mankind are right-handed, the British norm of driving on the left must be unnatural, hence the reason for the warning. Intrigued by what seems to be a self-evident anomaly, I check world statistics and discover that one-third of mankind drive on the left. But India, with 1,100 million inhabitants, is included in that total and thus makes the bulk of it, and since India was once a British dependency, it obviously must have been a British legacy there. But on closer inspection, I find that Japan and Thailand, which were never part of the British colonial empire, also adopted left-side driving. Why would this be? I check wikipedia, which is not always reliable but does usually have a jumble of informative material on any subject under the sun, and I find dozens of theories about left-side driving, among them that Romans might have driven their chariots on the left-side of their immense network of roads and that in Japan left-side traffic was adopted during the Shogunate to prevent samurai from unintentionally clashing swords, worn of the left, when passing each other. There is even the claim that, for various possible but unverifiable reasons, humans always chose the left side to the right side in passing each other. In other words, there is no universal consensus on left-side or right-side driving, but there are two facts: UK left-side driving and the London warning. The rest is speculation. Some of the speculation is based on speculative premises. So how can I explain why London has to warn people about where they should look for traffic? I have no explanation, because obviously it wasn’t done for the convenience of tourists. But there must be an explanation. In this essay we make a stab at the validation of speculations from a historical perspective. We argue that speculations or interpretations can only be validated from “historical consensus”. There must exist “consensual opinion or opinions” on why one-third of mankind drives on the left-side of the road. Perhaps the explanation about Japan is valid and then perhaps not. What criteria can we use to decide?

Propositional attitudes (beliefs, hopes, desires, et al) are related to types of propositions. In general, belief attaches to valid propositions and disbelief to invalid propositions. Belief attaches to perception, but perception can lead to false belief, as in seeing a cloudy sky and believing that it is necessarily going to rain. Cognition is the pre-condition for belief or disbelief. Cognitive processes determine the typology of propositions. A typology of propositions is implicit in the fact that in awareness we can distinguish between perception and thought. Even though we can think and perceive at the same time, one cognitive activity predominates over the other, which is fractional, and what the simultaneity demonstrates is the inter-activeness and speed of cognition. In the positing itself of basic cognitive propositions (perception, memory, intuitive logic), we imply a distinction between these and all other sorts of propositions. We can make further distinctions in the generality of non-basic cognitive propositions—derivations from basic cognitive propositions—from the simple understanding that basic cognitive propositions being different sets of propositions, the derivations or inferences from basic cognitive propositions must also constitute different sets. It is cognition in its entirety that can guide us in the matter of typing propositions, just as it is the act of thinking about thought that makes possible all ideas about cognition. A typology of propositions offers a “blueprint” to the way the cognitive system is built.

There are three types of possible inferences: (1) apodictic, (2) necessary, and (3) contingent or possibilistic, which are those based on justified but not necessarily valid premises or inputs. (1) From apodictic derivations we can establish the category of logical and mathematical propositions. (2) From necessary derivations we can establish the category of factual and scientific propositions. (3) And from the category of contingent or possibilistic derivations we can establish the category of probabilistic propositions. Inferences in themselves are of course sequences of propositions “molded” into logical forms. The typology of propositions is based on the results of inferences. Probabilities are ultimately based on facts. In science, they are known as hypotheses. In categories of thought in which factual propositions are either unavailable or insignificant, probabilities are interpretative propositions or interpretations. Logical and mathematical propositions have in common that they are based on axioms and rules of derivation. Their validation is implicit in the justification. Logico-mathematical derivations are not controversial if they are derived "nomologically", but it is possible to have interpretative meta-mathematical or meta-logical propositions. Factual propositions are such that they are validated by the cognitive processes themselves, but in the process of justification from basic cognitive processes it is possible for error to creep in. The prototype of factual propositions is perception. Factual propositions also include perception-based propositions. We do not actually need to perceive things to have a factual knowledge of them. The factuality of perception-based propositions is as "strong" as that of perception itself. Perception relies on memory for its continuity. If there are grounds for scepticism about perception-based propositions, then there would also be grounds for scepticism about perception itself, for “discontinuity” enters into both forms of factuality. We can make an analogy between the discontinuity of individual perception and the "distance" between our perception and the perception of others. Formal truths or perception-based propositions constitute an area of knowledge and an aspect of reality different though hardly separate from perception, and in our everyday lives we are aware of and can easily make the distinctions between them. The standards we apply are different. We look up a social fact but we look again to make sure of a visual detail; we expect our ability to perceive to be more reliable than our ability to recall events; and especially we know that we are even if all else is crumbling about us. I perceive, therefore I am, because in perceiving I am thinking and the thought in the cogito argument is such that it is unlikely for a child to have it.

Certain factual propositions are not validated in the process of justification, but they can be validated or not validated in time by facts. The relation between justification and validation is also applicable to such propositions, but it is deferred in time. We can describe these propositions as “deferred factuality”. The verification of everyday facts is a practical matter of their being reliable, even if their reliability is inevitably beset by “instabilities” of all sorts. We can predict that something will happen if certain precise conditions are met, but we can never actually be sure that something is going to be there when we go from one place to another in the course of our everyday lives until the act itself of perceiving. Scientific hypotheses are validated through predictability and are also a form of deferred factuality. There are two forms of the predictive scientific proposition. One form is the statement of scientific laws. The other form is constituted by the statements made on the strength of scientific laws. Scientific laws are alternative ways of describing everyday events such as objects falling, a plane flying, a light bulb shining, etc. They are generalizations about types of events, so that a heavy object becomes a mass, an airplane a surface across which air flows, an electric lamp a resistance, and so on.

But if we used the criterion of predictability, all descriptive and taxonomical scientific propositions would be excluded from the realm of science. If we invoke different criteria, we shall face ambiguity and un-specifiability. If deferred factuality, which is in essence not different from factuality, is the criterion for science, then factuality itself must function as a scientific criterion. And since descriptive and taxonomical propositions are methodically and systemically factual, we have only to distinguish between ordinary factual propositions and scientific factual propositions. To posit factuality as criterion means in effect that just perceiving would have to be considered scientific, but this is hardly the case because “indiscriminate” or “unsystematic” perceiving, one, can be erroneous, and two, is not necessarily reliably scientific. The answer to this argument probably lies in the qualifiers "systemic" and "methodical". Science usually involves a special form of factuality. It is not unaided perception. It involves repeated, specialized, and enhanced perception. What validates such propositions is the type of observation involved and the arguments on which such observations are based and the arguments which can be derived from such observations. All of these observations and arguments must be such that they resist refutation and that they provide sufficient and necessary explanations for the phenomena they refer to. If in addition to these properties such propositions also allow some margin for predictability, what we have is additional validation. In the case of biology, although initially purely descriptive (natural history), with some haphazard predictability, it eventually became explanatory and the possibility of predictability became stronger. But is molecular biology predictive at all? It is obviously predictive when its principles are experimentally applied and this of course is validating, but the principles themselves are purely descriptive. Science can be predictive in the sense of a form of deferred factuality, but it is more adequately defined in terms of systemic, methodical, and enhanced perception and of the inferences that can be justifiably derived from such factuality. Much if not all modern science has been validated from its "permanence" and from the imperviousness of such permanence to criticism or from its repeated validation in experience. Thus, predictability, which seemed so sharp, so final in its enunciation, must give ground as the criterion of scientific validity to factuality, and science cannot avoid the temporality of facts.

There is no logical principle that licenses the inference that one moment of perception necessarily follows upon another moment of perception, not even to say that it is necessary that one moment of perception should lead to another moment of perception. Logic does not included axioms or principles about time. How can I be absolutely certain that what I saw a moment ago actually happened, given that I no longer have the self-validating evidence of the event? The reasoning around any past proposition could be: “This happened or was happening an instant ago; it is not happening now, but I know that what I am seeing now is a continuation of what I was seeing an instant ago. How do I know this? I just know it, and no one would dispute that such is the case.” There is no question of perceptual propositions being merely probabilistic. Nevertheless, what emerges from this analysis is not a clear distinction between factuality and probability, but more and less complex cognitive operations, so that we could say that the difference between types of probability does not lie in the contents of the propositions but in the complexity of the cognitive processes involved, which can be deduced from the contents of the propositions themselves. Acts of perception and memory are complex processes whereas probability simply assumes both, as in my thinking that the building I work in will probably be standing when I get there. This, however, is a rather cumbersome standard. Perhaps we may find a more expeditious one. Cognitively, facts “best” probabilities. Probability must always involve factuality however marginally probabilistic facts may be. Does this exclude probability from probability? Not necessarily, but at some point probability must rely on factuality. Otherwise where would we stand? How could we say that a probable theory about the origins of feudalism involving a certain proposition based on the reading of documents, rather than on their factuality, will permit us to establish the probable validity of such a proposition? The circularity is too glaring! Probability must originate in factuality, such as the fact of the existence of a historical document. Since the factuality of perception extends to the factuality of a document or a recovered artifact, how can we distinguish between probability in the sense of a conventional definition of present and probability in reference to a remote past? In fact, no such distinction is possible, and what we are led to say here is that probability is instantiated in propositions which, although not themselves factual, must be based on factual propositions, assuming that a document we are using is genuine and not a fake. Probability does not usually apply to logical and mathematical propositions, although it is possible to speak of a probable mathematical derivation, but this only means that the full working out of a set of mathematical strings or logical formulations has not been fully worked out. In this particular sense, it is a special form of probability, for probability in terms of factuality leaves open the door to doubt whereas in logics and mathematics it implies incompleteness. There are, then, “degrees” of probability.

Besides apodictic inferences, factuality, and factual probabilities, there is the probabilistic class or type of interpretative propositions. The defining property of these propositions is that they are controversial in that they seemingly cannot be validated. It is only possible to justify them. We interpret history. Beyond historical probabilities based on facts, there are also theories based on probabilities and facts and these theories are interpretations of facts and probabilities, but not themselves either facts or probabilities. A theory on the origins of feudalism based on a probable inference from the fact of a historical document cannot be said to be anything but interpretative. Certain categories of social propositions are also interpretations of facts and probabilities, even when based on statistics. Natural history involves both factual propositions and past probabilities. Is the theory of evolution mere interpretation? This depends on the reference of "theory of evolution"? If it refers to the factual evidence for changes over time in the life-forms predominant on Earth, then evolution is a fact. If it refers to the likelihood that species evolve from other species, there is also little room for controversy. If it refers to the "mechanics" of evolution—whether it is reproductive success, or the survival value of certain organs or traits, or some kind of genetic process—then we are certainly in the area of interpretation and controversy, and as long as some conclusive evidence does not come along to settle the issue—which concerning natural history, as in the case of the extent of mass extinctions, is often quite improbable—then there will be controversy in this area of evolutionary theory. But there is no question of making natural history in general purely interpretative, for the simple fact is that all natural-history hypotheses must be based on factual propositions. In science, it is not admissible to build hypotheses on probabilities. However improbable relativity might originally have seemed, it was based on the verification of quantum theory. Despite some odd, faddish theories about science being “contextual”—the “Kuhnian paradigm”—it is generally agreed that science advances through a process of hypothesis-thesis and consensus-building in which interpretative error is a “natural” part which eventually gets side-tracked.

Are philosophical propositions in the same category as controversial historical interpretations and political and social theses? For the contents of the history of philosophy, as opposed to specific philosophical theses, there seems to be a wider basis for consensus. This wider basis could be identified as the recurrence of “fundamental historical propositions”, from the pre-Socratics onwards in the case of Western philosophy, in the three distinct branches of philosophy, that is, those implicit in the concepts of being (ontology and metaphysics), knowing (epistemology), and doing (ethics and history itself). This tripartite division of philosophy is based on the idea of recurrence. What recurrence seems to indicate is a fundamental consensus about what philosophy is about. But can it be said that recurrence eliminates interpretation and controversy in philosophy? All that recurrence permits in the history of philosophy is to propose definitions and basic themes. But within these themes (being, knowing, doing), all sorts of controversies are possible. Similarly, the history of politics also posits certain fundamental themes, such as the state, democracy, autocracy, and so on, and again within them all sorts of controversial interpretations are possible. Recurrence, then, does not solve the problem of controversy. It does lend support to the idea of “historical validation”. We have not made any conclusive statement so far as to whether controversial propositions can be more than just justified. When we invoke history in connection with knowledge, it is not really, so far at any rate, with the idea in mind of finding a fully reliable means of validation, although that is the way towards which we are pointing.

Can we limit controversial, interpretative propositions to areas of thought? Can we say that these propositions are to be found in such areas and only in such areas? If we did, then we could also say that what we are defining is not a type of proposition, but “areas or fields of thought”. But we cannot take that step because all areas of thought use all types of propositions. In taking such a step we could be skirting the “fragmentation” of the essential sameness of cognitive processes for all propositions. Interpretations can be derived from factual and probabilistic propositions. There are cases from the history of science in which a "commonality" of evidence leads to different controversial interpretations. By the end of the 19^th century, the existence of the atom was consensual, but its structure was the subject of controvertible theories as gradually it became clear that it was composite and that it was not gravity that held it together. If we cannot make strict matches between types of propositions and areas of thought, then all types of propositions are to be found in all areas of thought. This implies that some wffs in logics or mathematics are controversial, which by definition they cannot be. It would be foolish to say that all types of propositions can be considered interpretations. But if we go to the root of the typology in inferences, what we can say is that all types of inferences can enter into interpretations. Still, interpretations are controversial inferences and this sounds incongruous when transposed into the previous statement. It would mean that in one proposition you could have three types of inferences and that it could be partly valid and partly controversial, as in saying, for example, that because Plato used Socrates as the protagonist of his dialogues, Plato and Socrates had exactly the same ideas. Since in any complex proposition it is the "weakest element" that typifies it, such a proposition would be interpretative and we can indeed say that all types of inferences can enter into interpretations.

Since inferences determine the typing of the proposition, then it is correct to say that all types of propositions can be involved in interpretative propositions and, theoretically, cognitive processes would allow us to say either “yes” or “no” to interpretations. Assuming a “yes” were justified, then individual cognition would seem to be validating such propositions. But ultimately this would be an illusion because however many facts may be involved in interpretations, these still remain controversial. It could be that a “historical consensus” is possible. This would mean that “consensus” is what historically obtains in respect to controversial propositions. Assuming that a consensus validates, then we could also say that it is only from history that interpretations can be validated. History contains all possible types of consensuses but it is from history also that a “global consensus” emerges. Consensus is different from history itself in the specific sense of being a product of history, something that becomes possible from history. Consensus and history are inseparable yet different concepts. But why involve history in consensus? Why not just say “consensus”? The controversial nature of interpretations means that justification is not possible from individual cognitive processes and that “transactionality” must be involved. Transactionality implies an ongoing and public process of debate. And in saying "ongoing and public", history is necessarily involved in the search for “global consensus”.

We can conceive of at least five possibilities for consensus which are not necessarily global: (1) voluntary or gradual involuntary retraction; (2) “community of believers”; (3) collation and conflation; (4) “ideal context”; and (5) “show of hands” of "those-in-the-know".

(1) Let us suppose opposed interpretations A and B. If A or B leave the field voluntarily, we would have a historical fact. But a voluntary retraction does not lead to the proposition: "This interpretation is valid". The retraction itself could be another interpretation. We are assuming that a public transactional epistemic process can validate, but this model does not validate in any way or form. At most what we have is a re-inforcement or more justification for one interpretation. Alternatively, suppose we have debate X and sides A and B, in which B is wrong. How do we know in fact that such is the case? There is no single objective criterion to tell us. Not knowing that it is wrong, B will stick to its guns, and in fact B is justified and it could conceivably obtain a consensus. Or gradually B leaves the field: perhaps because A has better papers on its side, or because it gains institutional support, or because it commands a majority of adherents, or because it imposes in some other manner on B. Such "validation" between opposed interpretations is suspect. A debate on an issue does not necessarily lead to “global consensus”.

(2) Another possibility, one which widens the scope of debate, is a historical process of collation and conflation. Suppose we have different and opposed positions A, B, and C. However, it is possible that A and B, or B and C, or A and C, could "come to terms", and it is also possible for a fourth position D to embrace all other positions. From here, can we say that D means: "This interpretation is valid"? But this does not resolve the problem of the impossibility of validating interpretations from their bases or premises. D embraces A, B, and C, but, since all of these are interpretations, D itself must be interpretative. Although such a wider debate would be historical, it would be so only in a restricted sense, for the denial of D, which also historical, is not contemplated.

(3) A “community of believers” would seem to be a good basis for transactionality and consensus. But a community of believers presupposes a "local" consensus and there is no reason not to suppose—in fact it is almost inevitably to do so—that there must exist another community of believers or other such communities on the same subject or in the same field that hold different interpretations each of which will negate all other possible local consensuses. This argument emphasizes the limitations of the concept of consensus. Consensus can be a rational but "localized" phenomenon founded on "zero-discrepancy" about basic principles. It embodies or expresses "fixed positions", and this is of the essence in controversy. In addition, a local or communitarian consensus tends to discount change and the possible “degradation” of consensual propositions. Consensus can be more "intransigent" than individual interpretation because a communitarian consensus implies that the transactional process has already occurred, whereas individual cognition is always open to the transactional maneuver.

(4) The “ideal-context” approach is an attempt to reintroduce objective logical or rational standards into a debate about interpretations. But ideal-context is an “imaginary” concept. It assumes, besides “equal competence”, consensus on definitions and a “Pareto-equilibrium” of available information. But does this solve the problem? Even if such an ideal-context were possible—and it isn't, or at least nothing like it is known in history—we have seen that justification never suffices to validate interpretations. Since the validation of interpretations could be assumed to be consensual, this wouldn't necessarily be a problem, but ideal-context implies identical thought processes and this worse than imaginary is inconceivable.

(5) “Show-of-hands” is a mere “numerical” approach. Towards which interpretation does a majority of scholars in a field incline at any moment in time? "Show of hands" is a metaphor for the preponderant opinion of “those-in-the-know”. It assumes that in any field of knowledge there are persons more qualified than others. The fields themselves are defined by those-in-the-know. But controversy surges precisely between there are those-in-the-know, and aside from a show of hands in such situations—assuming equal competence, which is basically what those-in-the-know means—there is no objective way to choose between different conflicting interpretations.

It would seem as if none of the possibilities for limited “historical consensus” that we have enumerated entirely specify the process. Except for “ideal-context”, none can lay claim on “globality”. Their conflation does not carry us very far either, for “show-of-hands” by itself is at odds with all the other possibilities. This does not mean necessarily that they are all invalid or useless. “Retraction” is part of the process of global historical validation. It is not the only means. A global historical consensus must transcend all communities of believers. Collation and conflation is a mechanical procedure that is beset by its rigidity, although in some simple issues it can be used. Ideal context is an imaginary concept, hence per se controversial. But the concept of those-in-the-know is crucial to historical validation. What we have obtained here are examples of “partial” consensuses and not the global consensus that we expect from history.

Since validation is possible from basic cognitive processes except for controversies, to which only a “global consensus” can apply for validation, history itself, which embraces all possible consensuses, would seem to be the only means for the validations of interpretations. It is only a global historical consensus that can validate interpretations. We shall call this consensus vox historiae. The initial claim of vox historiae is that it is "history" that "chooses" between alternative interpretations. In the search for knowledge in the fields where controversy is the norm, it is not possible to establish a rule of thumb such as following public debate. It may even be necessary to admit that the concept of historical validation can only be specified from itself according to the terms involved in each specific debate, as in saying, for instance, that historical validation in the natural sciences is not the same as validation in history of philosophy. But this does not disqualify the concept itself. Historical validation stands, only we cannot allow ourselves to circumscribe it by a rigid set of norms to be applied in a mechanical or routine fashion in some areas or fields and not in others. Yet again, we cannot discard entirely the possibility of such global norms or principles, which if they can be teased or derived from somewhere it has to be from history. It could be said that the specification of vox historiae consists in the development of its implications. Vox historiae consensus about interpretations might seem to be related to a “consensual definition of knowledge”, in the sense that, for instance, that perception is coin of the realm in epistemology. But all we can achieve from this is reciprocal clarification. The consensus implicit in cognition is a product of evolution. There is no debate on the reliability of perception. Sentient beings have been perceiving and reacting successfully on the data of perception since the first forms of life learned to distinguish between stimuli. The consensus that results from vox historiae is similar to cognitive consensus only to the extent that it involves the use of cognitive processes, but it is to be doubted that vox historiae could achieve the authority and the reliability of the consensus that exists on innate cognitive processes.

When we say that interpretations are such that no amount of justification can validate them, it is understood that something is missing in the cognitive process. Let us assume that it is “transactionality”. Processes for the interchange and debate of propositions could be the vehicles for “transactions” leading to consensus and the validation of controversial propositions. Transactionality can be undertaken individually or it can be a collective process. Vox historiae can be defined by the set of propositions that describe the transactional nature of public, collective cognitive processes. No partial public, collective process can produce valid interpretations, although this does not mean that it may not do so. Some "local consensus" could be in possession of a valid interpretation. But history as "comprising" the “totality” of valid propositions is more likely to produce valid interpretations through a consensual process that supersedes any "local or restricted consensus". But public, collective processes of debate on interpretations do not per se lead to validation, basically because what are involved are interpretations and interpretations of interpretations or derivations from interpretations and these by definition do not yield valid propositions. In the search for the propositions that constitute vox historiae, we must have recourse to facts which can globally validate interpretations. Given the lack of a factual basis for interpretations, the valid propositions that we are looking for are not historical facts as such but valid propositions about history. A proposition can be thought of as the result of a process. A cognitive process involves many propositions. The parts of any cognitive process are propositions. In the case of individual cognition, we can fall back on basic cognitive propositions, but in the case of historical validation there are no such propositions. To sustain the vox historiae argument, we have to find the public, collective equivalent of innate basic cognitive propositions.

The present is an instant. It is a virtually imaginary concept. All is past and the individual past emanates from the subconscious processing of propositions. No sooner do events occur than they enter pastness. They are recovered in memory. But they are also recovered in oral or written expression. This recovery can take the form of historiography, but it can also take the form of all other means of transmitting information (files, documents, diaries, etc.). The recovery of the past is a variegated process in which historiography can be considered its systematic or methodical form. Within history it is possible to establish categories such as politics and economy. One of these categories is the public and collective search for knowledge, which can be of knowledge in any of the categories of historical events including politics and economy but also the search for knowledge in any sense. The recovery of the past includes the public and collective presence of political events but also it can be the impact of political events on the search for knowledge. Arguably, it was the political influence of Catholicism in Spain that relegated scientific research in that country to a marginal position. If Spain was virtually absent from the progress of physics during the 19^th century and beyond, it was because of the social system that was preventing it from doing anything other than assimilating what physics research was doing elsewhere. Whatever Spanish scientists were doing it was irrelevant to history as the search for knowledge.

The past is not an undifferentiated mass of propositions. The past is recovered in its representation (as in memory) and in its expression (as in historiography). The recovery of the past is done in discrete time frames. It is the recovery of the past itself that defines the fields of historical research. The discreteness of the past is one of the functions of historiography, although the discrete specification of pastness can also be done from "specialized" perspectives, such as politics or economics, different from a general historiographical perspective. If politics and economics can achieve the same cognitive operation in relation to the recovery of the past as historiography in general, would the recovery of the past by politics and economics not then be the same as historiography? Or historiography, for that matter, not be mainly about politics or economics? When we say of historiography that it defines the discreteness of the past, we mean that it is one of its specifically proper functions, and this cannot be said of politics or economics. Historiography comprises all specialized points of view. In all cases, the one thing that can be said unequivocally is that events themselves do not create their historiographical discreteness. The Renaissance did not know itself as such and the term to designate the period it covers only came into vogue in the 19^th century. The English “Glorious Revolution” was an idealized view that became current among historians with the benefit of hindsight, for the participants in it themselves could not foresee that it would lead to the advance of democracy during the 19^th century in Britain.

There is no precise correspondence between the concept of history and the external correlates on which it is founded. The concept of history is built up "synthetically" from events as they are “recovered” by historiography. Historical or historiographical thinking requires a greater "commerce" with "reality" than metaphysical thought, but reference to facts does not warrant the validation of interpretations. Getting the dates of the Crusades right does not guarantee that our views on the Crusades must be correct. The recognition of facts, such as government and the legal system, does not endow our politics with the condition or even the possibility of being valid. Both the history of the Crusades and the opinions on what they were about, are based on the same factual sources. Likewise, all politics admit the common basis of the existence of the state. In these cases, opinions build on the same external correlates that sustain valid propositions. Facts advert to external correlates, but so does interpretation. Historical evidence licenses valid propositions. But our politics are only valid to the extent that "history" itself "says" that they are valid. All historical events are, of course, knowledge in one way or another. They involve cognitive processes. Much, if not most, of this knowledge, including the knowledge of the events themselves, can be explained satisfactorily from individual cognition. The problem is the validation of interpretations. What we are considering in history as knowledge are the public processes for the validation of interpretations. And what we are trying to define are the propositions that we can derive from knowledge as a historical process that can serve to validate interpretations. These are the propositions that we call vox historiae. Controversy can only be resolved through consensual propositions that emerge in and through history. On what basis is consensus achieved? When is consensus manifest? How can consensus be broken and why? The answers to these questions are all pertinent to vox historiæ. It is not what I think or what you think or what someone else thinks that is valid. It is not alone a question of a show of hands, because you can have, among those-in-the-know, majority views that are wrong and minority views that are right. Vox historiæ as the criterion for validity involves a shift of emphasis from individual cognitive processes to the public, collective “processing” of propositions. The propositions that constitute vox historiae are specific, history-derived propositions on the aspects of reality as well as specific inferences from such propositions.

I grasp an interpretation. I cannot validate it from basic cognitive propositions alone. But if I take the propositions of vox historiae and apply them to the interpretation, I have a basis on which to conclude that the interpretation is valid or is not valid. Let us look at this more closely. To start with, when I grasp an interpretation I will probably also grasp its justification. I can either agree or disagree. Logic and reason—reason being the logical processing of empirical propositions—will not by themselves allow me to conclude that it is a valid interpretation. Now what happens if I apply vox historiae? Theoretically, I possess propositions that allow me to conclude that this interpretation is the globally prevailing opinion at this point in time about this specific debate—on the assumption that I do not misread the facts of its prevailing—and that it has been elaborated through the individual cognitive processes of those-in-the-know and that it has gone through the public, collective, transactional processes by which such propositions are examined. Is this validation? It may not be strictly so, but it is as near to validation on such matters as one can hope to get in any given historical circumstances.

About historical facts—that there was a revolution in France in 1789—it can only be said that history validates if there is room for uncertainty. History does not validate facts. History as facts is self-validating. Historiography collects the facts. But historiography and the other means of recovering the past also interpret facts. Only history as validating in respect of itself as the historiographical recovery of the past, can validate interpretations. To validate interpretations, we need propositions, and these propositions, from previous arguments, must have to do with or must be found in history. Since interpretations by definition are not validated by facts, these propositions to be found in history are not the facts of history in themselves. They must be valid propositions from and about history including the facts of history. In saying that such propositions validate, we are classifying such propositions as cognitive propositions. We know that such derivations cannot be basic cognitive propositions because these are innate. Therefore, the derived cognitive propositions that constitute vox historiae are inferences for validating controversial propositions. They are inferences in the sense of being derived from history but also in the sense of being derived with basic cognitive propositions.

Before factual history establishes the validity of interpretations, it chooses what issues can be the subject of public and collective debate. Interpretations are disciplinary. However fragile the premises in any debate, the debate itself is the first point on which there must be historical consensus. Such validation is at least a “global consensus”. It entails ground rules. The participants in controversies have usually had something to do with raising the issue and they are experts on the ground rules. They are those-in-the-know. Among the rules is that they control or have access to the venues of debate, which include institutions and media. This is a bit like buying something and paying yourself for it. It is a process tainted by circularity. But it is the only way and it is not unknown that external influences manage to penetrate the circle of the cognoscenti. A further unavoidable circularity is that it is those-in-the-know who determine who they themselves are. Since the debate is on-going, there is always space for cracks in these circularities. When we say that "history validates", then, we are referring to different categories of propositions about history: (1) the recovery of the past as the "presence" of public and collective events, whether as factual or as controversial, e.g., the existence of Egyptian queen Hatshepsut is well-documented but the life of Semiramis is enveloped in legend; (2) the recovery of the past as the search for knowledge in the past including knowledge of public and collective events and of itself, as that the German historian Leopold von Ranke considered that each period in history deserved to be studied in itself from original sources and he was in rank—just couldn’t avoid it!—opposition to Hegel’s totalizing view of history; and (3) the reciprocal influence of public and collective events on the individual search for knowledge and of the individual search for knowledge on public and collective events including itself. Galileo was hounded by the Catholic church but his famous statement on heliocentrism (“E pur si muove”, meaning the Earth) showed his confidence in that his beliefs were stronger than the papacy. Historical validation consists in specifying processes about the recovery of the past.

Historical validation can be based on the validation of propositions in the past and on the validation of propositions on the search for validation in the past. It can depend on the facts of history as they affect the validation of propositions as well as on the effects of the "present" or past validation of propositions on past or "present" historical facts or the becoming of such facts. The same analysis can be applied to any process in which events themselves had a decisive “say” in the way in which the past is recovered and interpreted. We referred to the French Revolution because of the clarity of some of its results, but we could also have chosen historical processes of whatever sort, including the global search for knowledge, in which events and their consequences did not necessarily yield propositions about themselves that can be thought of as incontrovertible. Lamarck formulated a coherent theory of evolution vitiated by his belief in the heritability of acquired traits or habits. Even without historiography, the events of history are reflected in the on-going process of history. If the methodical, historiographical recovery of the past yields some incontrovertible propositions—that World War I did not end all wars—whereas the recent recovery of the past—the historical interpretation of the collapse of the Soviet Union—does not yield similarly incontrovertible propositions, this does not mean that history does not validate, but that the validation of propositions about certain events involves the same process of multiple and complex interaction between events, their cognitive recovery, and the debate on their cognitive recovery, but on a higher pitch of controversy because of the circumstances of the events themselves or our nearness to them or of the manner in which they are recovered; thus, it is probably still too early to speak of the “final” stabilization of post-Soviet Russia, and certainly so of the “final” political shape of the Balkans.

The expression of the facts about history that constitute vox historiae can be described as "derived cognitive propositions". It is their application to cognitive processes that can lead to the validation of interpretations. Vox historiae can refer, then, to cognitive propositions and to the validation of interpretations as “historical consensus”. The process of historical consensus involves something like “state-of-the-art”. Evidently, vox historiae cannot be a consensus in the restricted sense of a "community of believers" or a "school of thought". “Consensus” is proper to communities of believers. Vox historiae presumably validates interpretation across all communities of believers. Since the vox historiae-perspective is constituted by historical facts about history, it is only marginally transactional as such, but it propitiates transactionality and emphasizes the transactional nature of the results of the public and collective processes. If we are to derive from history propositions to validate interpretations, these propositions should be global, flexible, transactional, and impersonal. They should apply as if an observer “in the know” were “outside” looking in. If we are part of the debate and happen to be left out of the agreement, we shall probably dispute that there is anything like an agreement. From the “outside” we can assume an impersonal view. From the “inside” it is virtually impossible to do so. Only from the “outside” can we say that a proposition is "emerging" as valid from among many competing propositions. What emerges at the end of the day is the proposition: "This interpretation is valid", or "This interpretation is invalid", assuming that any proposition that isn't valid is invalid.

Bearing in mind, then, the distinction between the facts of history in themselves and factual propositions about history (not excluding the facts of history), vox historiae can be said to be constituted by at least the following propositions.

(2) History contains all interpretations. These two statements are incontrovertible.

(3) It is through public, collective, and transactional debate that consensus on interpretations emerges in a specific temporal context. It is one thing to debate the Bolshevik revolution at its start and another to do so after the USSR collapsed.

(4) Since politics has a history of heavy-handed and misleading meddling in controversies, the validation of interpretations is usually but not necessarily "un-coerced". The validation of interpretations is often achieved not through historiæ rerum gestarum (historiography) but through res gestae (the facts of history). The end of World War II put an end to any possible debate about the merits of Nazism.

(5) In the historical debate on controversial propositions, the crucial participants are those-in-the-know. As there are many debunkers and contrarians, this statement could be disputed. But in general it holds. Unless you come up with a singularly brilliant idea (and even then), to be published on academic issues you usually require membership in an institution, recommendation, or some sort of related leverage.

(6) Historical debate implies state-of-the-art, although this should not entail the “arbitrary” exclusion of interpretations. Various persons were deciphering cuneiform at the same time during the first half of the 19^th century, but it was only when their translations were collated that it was possible to say that cuneiform had in effect been deciphered. Before that it would have been arbitrary to exclude any attempt at decipherment.

(7) The consensus that emerges from the debate on controversial propositions should be valid globally in the sense that it cannot be localized. This statement is easily the one that could provoke most controversy. The history of India has probably been more studied by foreigners than by Indians themselves, but no self-respecting Indian would admit that any “foreign” interpretation of Indian history can be validated.

(8) From the historical nature of interpretation, a validation cannot be considered "definitive". This proceeds from the fact that the validation of an interpretation does not change its intrinsic interpretative condition.

(9) Valid interpretations "emerge" in a manner “analogous” to the way cognition grasps and elaborates propositions. Perception “emerges” from the instantaneous “piecing” together of sensations. Vox historiae can be a shorter or longer process, but it certainly is not instantaneous and it never has the reliability of perception. Energy from fission is contemplated in Einstein’s special relativity equation, but it wasn’t until 1938, that Lisa Meitner, Otto Hahn, and Fritz Strassmann collaborated in the actual splitting of an atom.

(10) There are issues on which consensus is unviable. A vast amount of propositions in the sciences of geology and of paleontology are and will probably always be controversial.

(11) The historical validation of controversies is beset by the problem of regress, which is the Achilles’ heel of historicism and we will be considering below.

If these propositions were norms, they would be as interpretative as the propositions we want to validate with them. But we can make two crucial arguments here: (1) they are not strictu sensu interpretations but mostly factual propositions; (2) they are not only factual propositions, but, like scientific propositions, they are propositions to which those-in-the-know habitually recur to sustain their positions. If we posit a type of cognitive process which is only valid for a certain set of humanity, are we not infringing the tacit principle of the “universality of cognition”? Science is universal, but the use of scientific laws to derive valid propositions is not something that everyone can do. Similarly, on the issue of the validation of interpretations, it is evident that, in many cases, this is not something that concerns the generality of humanity, although of course such cognitive "means" as scientific laws and vox-historiae principles are open to all individuals who can learn to use them. The universality of cognition is predicated not of cognitive derivations, but of cognitive principles. The application of vox-historiae principles is a means to determine how different positions stack up in specific controversies. They are somewhat like basic cognitive propositions, but only somewhat. Basic cognitive propositions yield apodictic and necessary inferences. Memory is fallible but the yields of memory are as “inevitable” as deductions from scientific laws. The “rules of memory” are unyielding. But necessary inferences are valid. How can memory be valid when we know it is fallible? Memory is “infallible” only in the sense in which awareness itself is necessarily “irrefutable”. We always act on what memory tells us. Without memory we would be paralyzed. In the case of the propositions from history about history and knowledge, their expression is in a conventional public, communicative language. Basic cognitive propositions can also be expressed in such languages. But vox-historiae principles, unlike basic cognitive propositions such as those of intuitive logic, as an “aggregate” are in themselves not necessarily above debate, although their claims are mostly factual, and in particular the deductions that can be made with them about actual controversies can be and will usually be interpretative themselves. Even when they identify a real consensus, the fact that it is identified from derived propositions about history means that it will be susceptible to “degradation”. In stating these principles from history, it would seem as if we really have not advanced much in the way to a solution to the problem of error. What exactly have we achieved if anything?

We know what we perceive without having to remind ourselves that we are perceiving or without having to stop and consider whether our perceptions are reliable or not. We simply assume they are. In the process of historical validation, the emergence of consensus is not as routine-like and unremarkable as individual cognition. Even those “outside-in-the-know” are often hard put to recognize or point to an emerging consensus. Nevertheless, there usually are certain signs that a consensus has emerged or is emerging. In the lucid and un-condescending Quantum generations: A history of physics in the twentieth century (1999), its author, Helge Kragh, reasons his way through his topic, but now and then he stops to review the statistical evidence in journals and by other indicators for the rising star of quantum physics and relativity, through which he shows his awareness that science, even though it is valid, is also a matter of gradual acceptance by the world community of scientists. His book—published in the nick of time—is an admirable illustration of what we mean in this essay by those-in-the-know and state-of-the-art. When controversies, as often happens, are "bivalent", “passive acceptance” can be the signal of agreement, but, if only because of human vanity, this seldom happens. This is somewhat like the local consensus that we previously described as an A x B debate, but placed in a concrete historical frame. The most frequent and most face-saving signal of consensus is the absence of strong refutation. And usually the most gracious form of acceptance is not to say so, but the quote or the repetition of arguments, or simply the unqualified citation. One of the problems with consensus is that its emergence and consolidation often leads to neglecting the mention of its origins, or to relegating a thesis to insignificance. The historian Robert Drews points out that in the early 20th century the controversial thesis of a barbarian late-Bronze Age invasion of Greece was so pervasive that scholars neglected to say that it had been a relatively recent and highly speculative hypothesis proposed by Gaston Maspero. [1] There was a time when the thesis of the Belgian historian Henri Pirenne that Western feudalism in the Early Middle Ages had arisen in response to the closure of the Mediterranean by Islam was carrying all before it, yet today it is normal for this thesis to be dismissed in obiter dicta by students of the period.

That "history validates" can mean that the search for facts is the process whereby the recovery of the past yields different interpretations of the past among which, through the process of the search for historical consensus, some interpretations tend to gain more acceptance than others. Within history we can further distinguish between the political events and the record of political events. This is the distinction between history and historiography, between res gestæ and historiæ rerum gestarum. Historical facts themselves impose certain propositions that are accepted as valid within the informal and general search for facts in the recovery of the past. Dénis Diderot's encyclopedia contains both valid and invalid propositions. The entire work as such was “valid” in the historical sense that such an enterprise was accepted in his time as valid. But the ideology underlying much of encyclopedic thought was not of such general acceptance then or now as to be thought of as uncontroversially valid. The past “validation” of Diderot's encyclopedic effort was a historical process just as much as the controversy that surrounded Enlightenment ideas. One of the principal targets of the encyclopedists was feudalism as concretely manifest in the ancien régime. Ancien régime itself can be a controversial concept, but it has wide acceptance as designating the pre-revolutionary political and social order in France. The history of political events, and concretely the events during 1789, put an end to the ancien régime and the ideas it stood for. This is a proposition about history which history itself validated. This validation was achieved by history in different senses: (1) in the immediate recovery of the past in the past, with or without the methodicalness of historiography, for in the representation and expression of contemporaries it was known that, whatever its merits, whatever ancien régime connoted—whether feudalism and autocracy or autocracy without feudalism, or whatever else it might have meant—for all intents and purposes the ancien régime as a system of beliefs was over; (2) in the subsequent political events, which, if 1789 had not put paid to the ancien régime, those certainly did; (3) in the recovery of the past short of historiography itself, that is, in all forms of oral and written communications about 1789 and subsequent events; and finally (4) in the recovery of the past as historiography, for no one can dispute that whatever 1789 and subsequent events achieved or did not achieve, 1789 was a watershed in French history. In all these complex senses—none of which stands on its own in history but are all part of the historical process of historical validation—it can be said that history validates whether or not all the forms of validation are incorporated into historiography, in fact, even if historiography hadn't even dealt with them, because the history of France is a continuous process in which each moment bears the stamp of all preceding moments.

Within each discrete past that historiography or any means for the recovery of the past define, including specialized perspectives within each discrete past, interpretations arise, and with interpretations debate, and with debate the possibility of validation. From this schematic presentation, validation would seem to be restricted to what historians have to say about the past. But this is only part of the story, because historical events themselves can also validate or invalidate. Before the Second World War, Nazism was a "debateable proposition". After the Second World War, with or without historians' "verdict", Nazism was ended. It can hardly do to say that historiography and historiographical debate of themselves and only of themselves always determine the validity of controversial propositions. Beyond historical facts, historical thought shades into interpretation. How could Hitler have come to power? How and why was the Holocaust possible? Propositions about issues beyond facts can only be validated through vox-historiae consensus. The defeat of Nazism makes the denial of its tenets a valid proposition. The consensus of history is to be found both in the forces that defeated Nazism and in the historians working on the issues raised by Nazism as a historical fact. Individual cognitive processes usually cannot arbitrate debates about history beyond the facts themselves of history. In theory, historical consensus consists in validation among controversial propositions. But if we only look at what historians have to say we can miss the point of the process of historical validation. Politicians can be "ahead" of historians. Even today, Marxism, whose political influence worldwide is much reduced, is the theme of countless historical and philosophical theses. Yet these arguments also have the obvious implication that politics and politicians can be the upholders of invalid propositions which nevertheless have a place in historical debate. Not long ago dependency theory was all the rage in the study of the historical relations between developed and underdeveloped countries, not far in fact from an orthodoxy. Dependency still is a common view among historians and politicians in underdeveloped countries. But few state-of-the-art historians today would accept as valid an undiluted presentation of dependency theses, even though most of the major dependency thinkers are alive and well. What can account for the downfall of these ideas? There is no identifiable, over-all specific refutation of dependency theses that turned the tide or made all the difference. But it is possible to cite a long bibliography of valid but piecemeal refutations of the central tenets of dependency thought.

In philosophy we encounter the peculiarity that being wrong does not necessarily entail discredit. Even though philosophers have always pretended to be in the cutting edge of thought, it is not possible to make claims about the validity of any corpus of philosophical thought. At best, philosophy as a quest for knowledge is a process of obsolescence. Descartes posited dualism, but Cartesian dualism was submerged long ago by the progress of physiology and of the biological sciences in general. Locke espoused tabula rasa empiricism, but this inherently controversial view has been made definitely archaic by molecular genetics. Hume was concerned with the philosophy of science, but his theory of causality is meaningless from the perspective of technological progress. Leibniz's monadology is irrelevant to any neurological explanation of perception. Kant explored mental phenomena, but his psychological ideas—he claimed among other things that certain obviously empirical cognitive principles were innate—can be disputed by anyone today acquainted with experimental psychology. Yet all of those philosophers are the subject of endless disquisitions, some of which are reputed to be more "knowledgeable" than others. We have here, then, (1) the thought of the philosophers themselves, much of which is invalid, and we also have (2) research on the philosophers themselves, which can be "hierarchized" in terms of its "knowledgeability". Are we dealing with anything that even remotely approximates "truth"? And if the study of the history of philosophy can hardly be said to be about truth, aren't we here twice-removed from "reality", in a realm that requires propositions additional to consensually reliable individual cognitive processes? Aren't we in effect in the realm where "history passes judgment", and where, among other things, the historical "weight of thought" is sometimes greater than its explanatory value? Since almost anything that was said by philosophers in the past and is extant today can be the subject of serious consideration, why couldn't Aristotle and Sextus Empiricus be considered to be the equal of each other? Much of what Aristotle wrote is demonstrably false. Despite its obvious irrelevance, the philosophy of Aristotle has a greater "historical weight"—it is more original, more comprehensive and coherent, more influential—than the radical but presumably derivative scepticism of Sextus Empiricus. [2] Must we conclude that one set of propositions is more “acceptable” than another, hence more valid, even though both sets are formally equivalent and equally invalid? It need no come to that. In the case of invalid propositions, validation applies not to them but to propositions about them as objects of study or research. Historians of philosophy sort among invalid propositions not with the objective of validating hopelessly invalid propositions but with that of determining historical facts on such questions as the presence of Aristotle but not Sextus Empiricus in the Aquinian Summae. Among propositions about invalid propositions, there is the palpable fact that there are vastly more studies on Aristotle than on Sextus Empiricus or on Descartes than on Geulincx. [3] The acts of reading and classifying ideas are facts that convert artifacts into facts for the validation of propositions.

The degree of historical validation most be global but it can vary from field to field. The possibility of validating philosophical propositions outside of academia is remote, but within universities cognitive psychology and other non-philosophical disciplines contribute to philosophical debate. On the other hand, theorists can argue until they are blue in the face about “market socialism”, and this concept will be lettre morte unless it is carried out politically. Contrariwise, politicians in the Third World can go on promoting state intervention even as their national economies collapse, and the voices that oppose them, academic or otherwise, will be on the valid side of the debate however useless their strivings. Historical validation can also be defined in terms of geographical areas. What is “valid” for the Third World may not be valid for Europe. But the validation of vox historiae can only be significant in a global context. The history of the Second World War is about all sorts of events. The scientific results that obtained in that period and the political consequences of the conflict made gigantic contributions to knowledge in the widest possible sense. Was this a contribution of the war itself or of its historiography? The validity of the science has nothing to do with the history of the war even being part of the same historical process. The war itself certainly resulted in the practical refutation of a host of controversial issues. The historiography must consider this to be one of the main results of the war. Facts and historiography must totally coincide here. Conceivably, there may be significant areas of the conflict which still have not been covered well or at all by historiography. Even if historiography has not had time to get involved in certain issues—or even in the unimaginable case in which it does not get around to getting involved—the historical events are known as such and they in themselves, with or without historiographical judgement, have effects on history and specifically on the validation of controversial propositions.

One way to illustrate that historical validation can stand on its own without the judgement of historiography is science. History validates science in the sense that inductive hypotheses are a form of deferred factuality. Hence, history validates propositions that are not strictly factual. We do not really need to say that all of science requires historical validation, because many scientific theories are intrinsically predictive, but by an extension of history to include deferred factuality, we can argue that vox historiae is as pertinent to science as it is to interpretations of a non-scientific character. Scientific truth does not come into the world like a Pentecostal manifestation. Copernicanism did not impose itself without a struggle. Geocentrism was “functionally valid" for certain purposes, such as the calendric measurement of time. Tycho Brahe never accepted heliocentrism. It was only after a historical process of debate among those-in-the-know that Copernicanism triumphed. But astronomy is based on facts and the debate about its basic propositions in the end became irrelevant. And Copernicanism per se was not the only controversial aspect of the debate that it aroused in astronomy. The debate itself was also a subject of controversy. Was science to be debated on the basis of agreed-upon principles among authorities or was scientific debate to be carried on freely and without the dead weight of traditional authority? Since it was facts and not philosophical principles that were ultimately involved in the debate about heliocentrism, it was only a question of observing the debate among scientists themselves over time for validation to emerge whether the debate had become a part of written history or not. As a rule, scientific discoveries “naturally” involve consensus. Consensus arises among those-in-the-know after which it becomes accepted as truth. But consensus, in our times as in the beginnings of modern astronomy, is not necessarily immediate. DNA and the scientific discoveries underlying contemporary technologies such as miniaturization and fiber optics, have been accepted without much debate. But quantum physics, relativity, and plate tectonics, to cite but three examples, did not command instant assent. The physicist Alan Guth invented “inflation theory” to account for why the universe did not collapse into itself in the first billionth of a second after the Big Bang, and there has been a lot of empirical evidence for the thesis, yet he is still waiting for his Nobel Prize. However brief the lapse between the formulation of these hypotheses and their general acceptance enough time elapsed—and in the case of plate tectonics, the process has been relatively prolonged—to allow the case for the applicability of vox historiae to science.

Historiographical and descriptive-science and natural-history issues are often the source of much debate and consensus is not easy to come by, except when the facts speak for themselves. Discoveries such as Knossos and the remains of the Indus River Valley Civilization, or the evidence for the Viking arrival in North America, have been as indisputable as the decipherment of Turfan and Kuchan. [4] There is consensus on faunal extinctions during certain geological periods and on the evolution of dinosaur species from their emergence to their disappearance. But the causes of extinctions are controversial, as is still the hypothesis of the descent of birds from dinosaurs. There is general agreement on the lineage of man and the classification of hominids, but dates on this process were “bones of contention” until not very long ago. [5] Though recent, the thesis of the cataclysmic flooding of the Black Sea is near-consensual, but its consequences are anybody's guess. Anthropological studies, once dominated by "generalizers" such as J.G. Frazer, ceded to "particularists" such as B. Malinowsky, and today this discipline is often speculation à la Levi-Strauss or classificatory, especially based on kinship relations within extant tribes. On the dark periods of history, debate rages between archeologists and the partisans of the primacy of documentary exegesis. Vox historiae is sometimes eloquent by its silence, as when the debate on the origins of South East Asian civilizations simply faded from the self-evidence of its Indian sources. It goes without saying that on political and artistic debates the only valid pronouncement open to vox historiae is that there isn't one.

Since vox historiæ veers from individual cognitive processes towards history and history-derived cognitive propositions, it would seem to pull the rug from under the idea of an “universal consensus” about knowledge and to make thought hostage to change and instability. But this is a valid criticism only in the speculative sense that we could be led astray by facts or that the study of history could impose upon us arbitrary views. Can history be trusted at all? We are part of history and the individual cognitive processes that yield knowledge cannot be separated from the facts of history and their representation and expression. Vox historiæ depends in the first instance on the consensus about what constitutes knowledge. What is not likely is that we as individuals can sort out among controversies, although this stricture inhibits no one from believing the contrary. It could turn out that history itself is a futile process; hence, that any historical process can lead to error. But we can perceive that history has not so far resulted catastrophic for humanity. It cannot be proven that this will always be the case, but we have to assume that such will not necessarily be the case. What is likely is that propositions that posit the paramountcy of history in validating propositions pose as many problems as they solve. Since vox-historiae validation is admittedly subject to historical change, where does vox historiae stop? Theoretically, the process could go on forever, changing according to the state-of-the-art as long as facts do not put an end to it. If facts were the issue, the propositions would shift from interpretative to factual, but we can ask: is there such a cognitive process that starts as interpretative and becomes factual? Is vox historiae, in sum, hopelessly circular and regressive?

One of the difficulties that vox historiae raises as the criterion of valid belief is that it can tell us a great deal about the past, but very little about “present debate”, and this only once we have gotten over the problem of the delimitation of “present time” which we normally do as a conventional and variable short-term, rather than from the instability of a past continually in the process of creation out of an un-definable future. This leaves the seemingly intractable problem of regress. Basically what cognitive regress implies is that that we cannot "touch ground". In the case of cognition, we touch ground with "evolutionary, universal consensus". This consensus is what makes basic cognitive propositions reliable in themselves. Circular means that you are always back where you started. In vox historiae, we have a case of regress and circularity in which today's "facts" can be overturned by tomorrow's “facts”, which in turn will not be conclusive either, and so on. We cannot assume either that the cognitive propositions derived from the facts of history are non-regressive, because they do not modify or stabilize the developing and changeable nature of cognitive yields over time. Is validity in respect to interpretations always relative? If we must "read" history at each juncture to determine what propositions are valid and which aren't, we will be in a very precarious situation. Still, on the matter of interpretative propositions, we cannot appeal to the senses or to logic, and we must have a standard. Is it the issue that we must have or that we have a standard? Our claim for vox historiae is that we do have a standard and that the standard is historicist, even though we have to admit that historicism only allows us to make claims of relative validity. Since we cannot know for sure—partly because if we did we probably would not have to philosophize—we hope that by making all feasible intelectual exertions and going through all the correct motions, we shall be philosophizing well, although we know that "well" in the end is the sanction that only factuality can provide. When we make a statement about literature, we are not saying: "This is true"—except about the existence of the artifact itself—but: "We believe that this is a valid statement; one that merits recognition and acceptance". Our statement is subject to the test of time. And because of this, we also know that we are in a chain of infinite regress. If we cannot be absolutely certain about the temporal context of validation—since we never know when it is definitive, then it can never actually be authoritative—if all we know is that philosophy can only transcend in time through the exertions of brains that haven't the slightest inkling on whether their ideas will transcend or not—ten years from now only specialists may remember the realism/anti-realism debate, but will any part of its contents qualify as valid?—then we have rational grounds for questioning vox historiae as a reliable criterion for validity. [6]

The relativity of validity is also expressible through the idea of relevance. What Aristotle said about the polis is probably still valid—where it is not a matter of formal or probable truth—but it is less relevant, hence "less valid", than what we believe about capitalism. Even as we question vox historiae, we know that the validation of interpretative propositions does not have to be eternal—its “globality” being also assumed—but if we reject vox historiae, what we are left with is babble. In history at least we have some ground on which to stand, although we should not expect it to provide us with the perfect measure for views that are continually outstripping themselves. If we retain the vox-historiæ concept but admit that we have no guarantee about the validity of its validations, that we really cannot affirm that the propositions on whatever issue that obtain historical sanction are correct, then all that we can claim is that certain propositions are “current” or are “acceptable” or are “prevailing”, or whatever equivalent expression we choose, but not that such propositions are “valid”. Ergo, we have no theory of knowledge—except a negative theory—because we do not really have any sustainable criterion for knowledge other than verification. Yet what guarantees have we got about scientific hypotheses now, which are admittedly a form of knowledge? The mathematical “string theory” about the universe or universes has adepts, but its validation is so problematical even its proponents admit that they work on it because they enjoy doing it.

Vox historiae is inevitably regressive and at any time in history it is exposed to the possibility of error. One way around this problem is that error is a necessary part in the historical process of knowing and that vox historiae involves not regress but the “degradation of knowledge”. Until 20^th century physics proved otherwise, it was believed that Newton’s laws were universally, and eternally, valid. They still are valid of course but no one accepts the view that time and space are distinct absolutes. But this still does not allow us to touch ground. We can not be sure at any moment in time of the validity of interpretations. Ultimately we might be on the wrong track in areas where there is no possibility at all of making final claims on validity. The usual attitude towards interpretations is to stand our ground until shown to be either right or wrong. Vox historiae is regressive and vox-historiae validation is temporal and relativistic. There are historical junctures when "debate" consists in the historical facts of war. The results can be indecisive, as during the 16th century wars of religion, and it can be "decisive" as in consequence of the defeat of Nazism. But as close to utter defeat as communism may be, the historical "debate" is not closed. In most fields of knowledge, humanity has been much longer in utter obscurity than in the relative enlightenment of "our times". Since knowledge has emerged from obscurity and error, then the conclusion is inevitable that error is part of the historical acquisition of knowledge. Thus far so good. But to believe that this process is "valid" implies, however much we might try to avoid the subject, something akin to a “philosophy of history”, a set of propositions, embodying perhaps “historical patterns and tendencies”, that would validate vox historiae-validation itself. Assuming that this point of view is defensible, what would be achieving? It would not put paid to regress, but it would be coming as close to that objective as possible. It would be at least a partial validation of vox historiae.

Notwithstanding our intention to restrict the use of history on the issue of interpretations to what history can yield in the way of valid propositions that can be used for the validation of other propositions, the fact remains that the propositions that we derive from history, no matter how individually factual, in the aggregate conform something not entirely unlike an interpretation of history. In addition, the underlying assumption for the set of principles is that their application is ultimately valid if the processes that they describe do yield valid propositions, and for this to be so history itself must be "valid". Any claims of this sort imply interpretation and philosophy of history. The principles are individually valid but only in the context of the greater validity of the historical process overall. It seems then that we cannot escape the theme of philosophy of history. The anticipation of the "movement of history" is the only means to put an end to interpretative regress, even though we have to admit that any theory about history will eo ipso be considered interpretative and debateable. Nevertheless, if vox historiae is to work in the way we want it too, then history must somehow be rational and "progressive". From a purely cognitive perspective, the historical process cannot be coercive in the way that the communists imposed T. Lysenko and the Nazis their racialist doctrines in their respective totalitarian societies. For vox historiae to work, history must be an open process. We must "know" that it cannot be an exercise in futility.

To round out our theory of knowledge we need, then, a philosophy of history, and not any philosophy of history, but a rational and progressive philosophy of history, one that could argue from historiography for certain values, for certain rational and humanistic ends. Leibniz and Kant saw the need for just such an interpretation of history. We cannot argue that our theory of knowledge necessitates a philosophy of history and let it go at that, because we would be going about in circles. Can we justify the sort of philosophy of history that we need? However we approach it, the "movement of history" is per se controversial. But we need not approach it from a vast, "predictive", world-historical perspective. In doing a theory of history, in other words, we need not engage in historical "philosophizing" à la Hegel or Spengler or Toynbee. We can be quite comfortable with C.G. Hempel's arguments against the possibility of historical laws or with A. Danto's thesis on the meaningfulness of history on an analogy with the "rational enchainement" of events in narrative forms. We can live with this, but we can also try to go a little further without being caught out in pseudo-philosophical rantings. One simple illustration is that if history is caused, if it is a chain of causes—and how can this claim be refuted?—then it must be rational, for it is reason that reveals successiveness and successiveness is the sine qua non of cause-and-effect. We are not overstepping here the bounds of what we can deduce about intuitive logic, for we are not claiming innateness for the cause-and-effect relation, but derivation from innate principles. Beyond the rational description or explanation of history, we can make moderate, well-reasoned claims about the possible course or about the possible tendencies of history. In sum, since we have argued for a knowledge/history relation, then we have to argue for the “rationality of history”, or else we must jettison much of what now we consider knowledge. Does this mean that we would be using interpretation to validate interpretations? Only if the facts of history can be considered to be interpretations. The rational reading of history would seem to be going beyond the facts of history, but only if we pretend to have an impossibly strong, predictive vision of the future. It is possible to define history as the becoming of being. Knowledge has been increasing over time and it will go on doing so. The becoming of being means concretely that "knowledge" is throwing ever more widely its expanding net of categories and that, within these categories, it is increasing the number and power of its explanations. We have no reason to believe that this process will not go on.

Ultimately, history as such "does" nothing. Certain facts of history allow for certain conclusions about the elaboration of propositions. Despite the precarious nature of vox historiae, it does permit inferences about the validity of interpretations and these inferences are as close to a process of validation as we can get. But in the final reckoning can any approach to interpretations ever be "sufficient"? The answer we fear is no, not even when we throw in the crown of historical controversies, which is that history has "meaning", not just in the narrative sense, but also in the sense of being about the inevitable progress of mankind. Are we perhaps treading here some teleological-theological mine field?

Are our religious beliefs "knowledge"? If we merely say "I believe in God", we are expressing knowledge in at least two senses: that we "know" this and that another person can say that I claim to "know" this. But the proposition of our belief in God is not verifiable knowledge. We would have to make explicit the cognitive processes involved in this proposition. We would have to "externalize" our internal belief. We would have to appeal to cognitive processes to justify and validate such a belief. And this is where the real vox-historiae problem comes in, more so than with philosophy of history, which can be bypassed as some justified optimism about science. Religious propositions are neither particularly reliable, nor subject to predictability, nor probabilistic, which presupposes some formal truth on which to base inferences. Consensus is a localistic, relatively narrow concept. The historicity of consensus gives rise to the concept of vox historiae, which can be defined as an expanded consensus and can serve to validate certain controversial propositions. Since consensus or vox historiae as an expanded consensus are legitimate validations in history and in the hypothetical or experimental phase in science, it can safely be assumed that it is also applicable in the case of religious propositions. But religious propositions are grounded on cognition and on affects. We are not dealing with the evidentiary value of formal truths as in science and history and we can only appeal to consensus; but since there are believers and atheists, it would seem that we can talk of two perfectly legitimate consensuses; in other words, there is no vox-historiae judgement. There is no way to validate religious propositions of the sort mentioned. And if that is the case of religious propositions, a case can be made that it is also the case of philosophical propositions, as regarding, for instance, the abyss between analytical and continental philosophies. Is there a way out of this impasse? One conceivable way out is in the concept of justification. Justification itself is a internalist view of knowledge. But justification cuts both ways, and therefore, we are stuck with the proposition that both the affirmation and the denial of the existence of God are forms or manifestations of knowledge. Since it is unlikely that believers and atheists can come to some kind of compromise, again it would seem as if there is no possible way to achieve vox-historiae consensus except, in the case of believers as to the arguments for the existence of God, and in the case of unbelievers as to their arguments. Again, consensus is a restricted, localistic concept. Since vox historiae is a set of propositions that permit an expansion of the concept of consensus, it is to be doubted that it could help us here.

In sum, there is no way that we can escape the problem of regress with historicism. But there is no way that anti-historicism can actually validate interpretative propositions. Anti-historicism breaks down into sects and cabals and local consensuses. Historicism at least can have a go at “relative validation” which is a few notches above cabals and sects. Even though it may be that there are indeed valid as against invalid interpretations at each historical juncture—and this is not necessarily so either—we must recognize that we really cannot know, even assuming all conceivable transactional processes, which are the valid interpretations. If none of the facts (all forms of collective elaboration, communication, and interchange of interpretations) and all individual approaches to validation (show of hands and communities of believers), all involve interpretations and none can really validate interpretations, what is the use of vox historiae? Must we give up on vox historiae? Is there no way whatsoever to "reinforce" individual cognitive processes from the overview afforded by a general historical perspective? Even though an answer on vox historiae and the validation of interpretations has to be: "No, it does not allow us to validate interpretations categorically", the use of the propositions in vox historiae does take us as close as we can possibly get in that direction. But there is an interesting implication here. Assuming valid interpretations but no means to identify them, vox historiae accommodates all interpretations except itself, for it is undeniable that history makes room for all possible controversies. Any specific theory about the validation of interpretations would exclude all other interpretations but not vox historiae. This makes vox historiae virtually irrefutable. Philosophical thinking is suffused by circularity, except when its objects fall into other categories of investigation. Cognition is the object of scientific study. Physics is the scientific equivalent of ontology. As a knowledge in itself, history is not the object of study of any particular science. There are no qualifications on the philosophical understanding of history and by default philosophy is quintessentially about history.

Ah, yes, concerning the warnings on London streets, I enquired dutifully but could not obtain a consensual answer, which meant I would have to do a lot of archival research to find the facts. I suggested they might have originated during World War II, when in England there were as many or more American soldiers as British ones, and that rung a bell in some people, but some one else said that there were those signs before the war. I finally decided to drop the subject as too time-consuming and fell back on my original surmise that driving on the left side is contra natura. But if there were no facts on who and when and why it was decided to mark the streets of London with these warnings, there conceivably could arise a debate on the issue and I knew I possessed the tools to decide if a historical consensus was emerging or could emerge.

[1] Robert Drews, The end of the Bronze Age (1993), pp. 53-72

[2] Sextus Empiricus lived in the 2^nd and 3^rd centuries CE. His principal work is titled Outline of Pyrrhonism. Pyrrho was a Hellenistic philosopher who lived between ca360 and ca270 BCE.

[3] Unlike Descartes, who believed in free will, the 17^th century Flemish philosopher Arnold Geulincx turned his back on Cartesianism by espousing total divine determinism.

[4] Both are extinct languages once used in the Tarim Basin (western China). It helped that there was a rapid consensus on that both languages, despite being discovered in the heart of Central Asia, belonged to the same family as most European languages.

[5] In Bones of contention (1987), Roger Lewin describes how, as paleoanthropologists discovered more and more fossils of potential ancestors of man, they tended, first, to classify them as failed species, and then, deliberately to place the apes/human split as far back in time as possible.

[6] “Realists” believed that perception is “logical”; “anti-realists” held that “justification” was equivalent to “validation” where validation was possible.