Again about history as science

Discussion on H-History-and-Theory list, June 2006xi

If madness lies in the position that science is an ideology, danger lies in the false distinction that what science “is” is separate from what scientists “do”. This position suggests that there is, in fact, an “ideal,” Platonic form that represents science, distinct from the activity of conducting science. There is, in reality, no such distinction. What science is and what scientists do are the same because, in order for science to happen, someone must do it. The attempt to make this distinction, moreover, is evidence that science retains an ideological fidelity from those who see a difference between the “ideal” of science and the practice of conducting science.

“Belief” is also a dangerous word. Invoking it leads to binary thinking, putting those interested in this question into two camps: the scientists and the “believers”. Such a distinction is a caricature of the issue, suggesting as it does that there is a clear and distinct difference between matters of fact (rationality) and matters of faith (irrationality). We do “believe” in science insofar as we have “faith” in its method, its promise for solving problems, and its (often counterintuitive) results.

Another caricature is to draw analogical similarities between people on LSD and those who do not show absolute deference to the rational activities of science. It is an intellectual cheap shot and adds nothing to the conversation. Furthermore, it is a rhetorical move which asks the audience to create an association between users of hallucinogenic drugs and those who are interested in understanding how science gets appropriated for un-scientific ends, ends that are usually couched in ideologies—ideologies sometimes held by scientists themselves and sometimes by those who simply want to reap the rewards of technology and sometimes by both.

Finally, the notion of ideologies as “closed systems” is a rather naive view of how ideologies come into being, how they are maintained, and how they evolve. In fact, ideologies do “admit” new forms into their acceptance frames all the time. If they didn’t, we would still hold the same beliefs that we did 100, 1000, or 10,000 years ago. Ideologies are “organic” in the sense that they do grow, transform, and take on new significance through the lens of dominant culture (in the West, this culture would have science as a main epistemological anchor) and residual cultures that are part of our collective history. (I’m not sure how championing Vitamin C is an “ideology”—especially since it was the pet project of a single man).

We talk at odds about this question, I “believe,” because we set up dichotomies that limit the range of possible discourse. It is counterproductive to look at this as a “two camp” argument, with the rational scientist on one side and the irrational believer on the other. There is much scholarship on the ideological uses of science, and it rightfully complicates the simple positivistic impulse to reduce this to an “us and them” question. I will be happy to provide bibliographical references to those interested.

David Tietge

In response to several contributors on this topic:

It seems to me that the first question a historian should ask is this. If a story is to qualify as history, will it be sufficient that it is a true story about past events, or must it also be story that is explicitly or implicitly explanatory of the events that it describes?

If truth is considered sufficient, then how shall the historian's practice be distinguished from that of the responsible journalist and the diligent criminal investigator?

I take it that satisfactory history stories must be explicitly or implicitly explanatory. There will still be ways of differentiating between historical styles. For example, the explanatory umbrella may be conceived by some historians as requiring the parade of evidence concerning the will of God, or of The Spirit of Progress, or as or as the outcome of a battle between Good and Evil; or as the evidence that demonstrably erratic and capricious demons are at play.

A consensus has grown in modern times that explanatory accounts of past events will be satisfactory to the extent that they draw upon shared suppositions about notional regularities that are believed to be deeply inherent in the universe. It is characteristic of such suppositions that, other things being equal, events occurring in one place at one time will be assumed to follow a course that is significantly similar to the course of similar events (other things being equal) occurring at another place and another time.

This schema (that might be very loosely characterized as the 'scientific’ world view) is, as stated, subject to one very strong challenge. There is often one respect in which ‘other things' can never be regarded as equal. For example, the second time we are offered a bargain by a plausible crook we shall be less gullible than we were. Evidently our notion of the equality of ‘other things' must somehow take time's arrow into account.

’Evolution’ is the general name we give to certain regularities that we do belive not to be absolutely or ideally universal, but to turn significantly upon the specificities of time and place.

It emerges that the sorts of things that can plausibly be regarded as responsive to absolutely universal regularities are particulars, while the sorts of things that can plausibly be regarded as regularly responsive to the peculiarities of their time and place are kinds. A bee does not evolve, and neither does a flying machine; but the bee evolves, and so does the flying machine.

To be a historian, I suggest, is to construct stories about the past that are both true and explanatory in two distinctively different ways. They will draw for some of their explanatory power upon the universal regularities that account for the vicissitudes of particulars, and upon those very differently structured regularities that account for the evolution of kinds.

Prediction is much less reliable in the explanatory mode appropriate to kinds than it is in the explanatory mode appropriate to particulars, but on those occasions when predictions are confounded we should always be able, in retrospect, to convince ourselves that we can see why.

Donald Brook

Scientific History—the science of final causes

In the mid-20 th century, history disciplines experienced the analogue of the 17 th century revolution in the natural sciences; i.e., challenges to prevailing philosophies and redefinitions of knowledge. Historians such as Lee Benson, Samuel Hays, Robert Fogel, etc., rejecting “narrative” historiography, promoted history based on scientific concepts (e.g. classification), methods (e.g. hypothesis testing) and techniques (e.g. quantitative analysis). However, unlike the 17 th century, this would be historiographic revolution has largely failed. Today, scientific historiography is a small sect in the history profession and in danger of extinction.

The failure of the scientific history revolution is essentially a failure of philosophy. Seventeenth century natural scientist did not simply reject prevailing ontology and epistemology; they created new philosophies to justify their new sciences. For example, Galileo, Newton and Descarte, did not just report the results of experiments and computations. They wrote philosophical treatises challenging assumptions of natural philosophy and developing a conceptual system for modern science. Conversely, twentieth century scientific historians, lacking the tincture of creative philosophy, while rejecting traditional narrative history simply copied the philosophy of science and attempted to apply it to the objects of historical enquires. Thus, it was destined to fail because entities of historical investigations (societies) are different than scientific entities (materials). Also, the cause of social change is different than the cause of changing matter.

The purpose of both modern science and pre-modern natural philosophy is to explain observable change in nature in terms of its causes (i.e. what causes things to change). Aristotelian science is based on four causes (explanatory principles) of changing things: (1) material cause (e.g. when heated, metal material changes differently than plastic material); (2) formal cause (essence of the thing—e.g. dogs react to same stimuli differently than cats); (3) efficient cause (the primary source of change - e.g., the carpenter changes wood into a table); (4) final cause (purpose or goal- e.g. health is the cause of exercise). Science has largely integrated the first three causes into its conceptual system. However, final causality has been categorically rejected as an explanatory principle.

Scientific historians similarly have accepted the first three explanatory principles and ignored final cause. The first problem with this blind acceptance is that the essence of society is not ‘material’, it's ’ideological’. What binds masses of individuals separated in space and time into a social entity (e.g. states, religions, etc.) is a common ideology (e.g. “Deutschland Uber Alles”, Holy Trinity, etc.). Secondly, recognizing the fact that the essence of human decision making is largely based on projecting ideas about the future, final cause has to be part of a scientific historiography. Humans project the future (i.e. set goals) and their behavior is caused by (explained in terms of) that goal. In short, it is impossible to explain changes in society without considering the ideological cause and final cause of social change.

While the empirical research of scientific historiography should and must continue along the lines exemplified by mighty historians like Benson, Hays, Fogel, etc., there is much need for ontological (nature of social being) and epistemological (nature of knowledge of social being) work. Like 17 th century natural scientist, scientific historians have to “wear to two hats”— empirical researcher and philosopher.

References:

1. Falcon, Andrea, “Aristotle on Causality”, The Stanford Encyclopedia of Philosophy (Spring 2006 Edition), Edward N. Zalta (ed.), <http://plato.stanford.edu/archives/spr2006/entries/aristotle-causality/>

2. Cohen, S. Marc, “The Four Causes” http://faculty.washington.edu/smcohen/320/4causes.htm

3. “Great Books of the Western World”, R. M. Hutchins (ed), vol. 2 pp.155 ff., pp. 193 ff.

Tom Verso

If any of you take the trouble to read a scientific paper you might see that it takes two forms:

(a)An experiment where what was done and what resulted is followed by what it means in light of other experiments

(b) An account of what the artefactual traces indicate has happened based on some rules of the meaning of artifact— a sort of geological archeology.

Now if that's not history, would someone please be so kind as to explain why to this old fool who spent his whole career diagnosing and trying to cure on the basis of “history taking.” By the way, if you look at any medical book, it is nothing but a history of what was seen to work and what that means (pathology).

DE Teodoru

Glad to see that my last posting produced so active response.

Responding to Steve,

1) Euklid geometry ( one of the basics of modern science), which I hope You did study in school based on 5 axioms. One has to accept this axioms without questioning. All other geometry was going from there. Suddenly in 19 century 4 matemathicians rejected to accept fifth axiom and created so called non-Euclid geometries. This new geometries also based on axioms but slightly different than Euclid. And one has again to accept in them without any proof Does not acceptance of something without any proof “belief”? Another fact , there is some evidence ( A. Koyre, A. Einstein and P. Feyarabend) that Galilei did falsify results of his experiments because he believed in his hypotesis. If science is providing most rationalistic explanation of surrounding reality and life at the moment from Western point of view ( in wide meaning of the WEST including Russia and Japan ), it does not mean that science is not ideology ( system of theories and related ideas). Suggested further reading for You ( if You are interested): “Against Method” by P. Fayerabend, “Straw dogs” by J. Gray and “ Strange histories” by D. Oldridge. You do not have to agree with authors ideas but they could show You some new perspectives. It is not madness, it is just a bit sobering idea for You.

2) I did not discuss impossibility of finding documents or other artefacts. I did respond to example which Haines did use. May be Haines example was unfortunate, but that is not what You are talking about here. If example was bad ( which I do not think so in context of discussion), blame example not response on it. And discussion of possibility of finding documents or artefacts which did previously exist or did not exist is totally different discussion ( different types of predictions, one could easely see difference. If one wants!)

Now, more seriously about Haines response.

1) I totally agree with You about that “science is an ambivalent word”. And that's why at the start of my participation in this discussion, I did express my doubts in possibility of definig history without agreeing first on what we are meaning when we using term “ science”. As a result, it could looks as everybody agrees when actually everybody will be using her/his personal understandig of “science” and as a conscquence “history”.

2) As Your example of Bolivia goes, my answer still the same. It is wild guess, if You want with high probability of success but still wild guess. One would produce this guess by evaluating economical, political, demografical and so on situation in this country. So one does not have to be historian or even now history to produce this wild guess. But You are saying as well:

“Particular outcomes such as you bring up, were not unequivocally predictable of course, but if we knew enough about prior situations, we hopefully would be able to know that such outcome was possible and even assign a certain probability to it. Without using this probabilistic prediction, the course of history would necessarily appear chaotic.”

This two phrases are another attempt to use modeling to predict future occurence of event. So wide cryptoreference to theory of probability suggests to me, that You instead of history discussing how better to apply chosen by particular researcher facts to predict future. Again, nothing wrong with that. But where is the history? This is modeling of future ( futurology).

And below:

“Except for the chronicer, don’t all historians explain a particular outcome in the past as a result of its prior situation including the intentions of the actors?”

Here is the history. Something happenned and we trying to explain why and how, and what does it mean for present if anything. Post factum, not before something occurs!!!

There are some additional thought in regards to some other points which You made in Your reply. But I think that our differrences in opinion clear enough. And my response is way too long.

Thank everybody for their thoughts.

Iaroslav

Iaroslav,

Thanks for your challenging reply.

> 2) As Your example of Bolivia goes, my answer still the same. It is
> wild guess, if You want with high probability of success but still
> wild guess.

I don't know that we are disagreeing here. You don’t seem to object my suggestion that Bolivia will probably be here tomorrow, although you label it a “guess”. I don’t much care for that word here because it implies an opinion without knowledge or means of knowledge or a random judgement. A suggestion that Bolivia will be with us tomorrow does not seem a random judgement, but one based on the knowledge that it has persisted many years, is well institutionalized, etc., and so I feel I can project its existence to some degree into the near future. You don’t seem to disagree on the substance of the point as you follow it with:

> One would produce this guess by evaluating economical, political,
> demografical and so on situation in this country.

> So one does not have to be historian or even now history to produce
> this wild guess.

That is correct, as my little elaboration illustrates. I suspect the underlying issue is that history is emergent, and so to that extent can’t be predicted, but to the extent it represents continuity, it can. But the plot thickens:

> But You are saying as well:
>
> “Particular outcomes such as you bring up, were not unequivocally
> predictable of course, but if we knew enough about prior situations,
> we hopefully would be able to know that such outcome was possible
> and even assign a certain probability to it. Without using this
> probabilistic prediction, the course of history would necessarily
> appear chaotic.”

We probably don’t disagree that if we knew what was possible and what was impossible about some situation, we could predict what won’t happen. But my point wasn’t about what was possible or impossible, but that there are forces at work in that situation (if we manage to grasp it in dynamic terms) that allow us to determine as well the relative probability of the various possible outcomes.

Perhaps I should not have spoken so abstractly, and so here's an example that is somewhat more specific:

In systems theory, it is known that when a system “matures” (contradictions deepen), it's behavior changes, and it “locks in”. Although it achieves a high level of empirical development in specific and narrow ways, it looses flexibility and scope for innovation beyond that narrow definition. So as a system matures, the probability distribution of its possible outcomes narrows. By knowing the “age” of the system, you know to what extent you can “guess” what will happen next. In terms more familiar to historians, this is called “structural determinism”: the behavior of a society's individuals is constrained by the character of the system in which they happen to participate.

> This two phrases are another attempt to use modeling to predict
> future occurence of event. So wide cryptoreference to theory of
> probability suggests to me, that You instead of history discussing
> how better to apply chosen by particular researcher facts to predict
> future. Again, nothing wrong with that. But where is the history?
> This is modeling of future ( futurology).

Yes, if by “modeling” you mean representing a static system in terms of an analog. In that case, prediction is possible. However, if we instead start with the assumption that every part of the system is actually a process, then prediction at best probabilistic, and if we know the age of the system, we can estimate the distribution of the probability of possible outcomes.

I guess it's time for a concrete example, drawn from a less contentious subject than historiography.

Suppose we have a supersaturated solution of a salt. Given that it is supersaturated, molecules will probably bond, but there are often several ways they will do so with differing probabilities. Knowing the features of the molecules, we know the probability distribution of these possible molecular bondings. However, from these initial possibilities, some type of bonding does establish itself and results in what is called a lattice structure that determines how most future bondings will take place. The probability distribution of possible outcomes (the shape of the crystal) has narrowed once a lattice is established. The lattice structure constrains the probability distribution of possible futures.

We have another example after this point as the crystal grows, for it reduces the saturation of the solution, and so the system (including the crystal and its environment) “ages”, ultimately reaching a point at which the probability of bonding equals the probability of dissociation, and crystal growth virtually stops. We could take an empirical measurement of the saturation of the solution at any point in the process and derive from it an idea of the system's age and its future potential based on that estimate of its age. If the saturation is low, we know that further development will slow and soon stop (reach equilibrium).

In historical systems, development necessarily reduces the potentials that had made it possible in the first place, and the options for possible outcomes narrow. This is what I meant by ascertaining the probability distribution of possible outcomes. We know that the capitalist system won’t revert to feudalism, and people acting within it find their behavior constrained by that system so that they either try to perpetuate the system (evolution) or seek to restructure it in some novel way (revolution) .

> And below:
>
> “Except for the chronicer, don’t all historians explain a particular
> outcome in the past as a result of its prior situation including the
> intentions of the actors?”
>
> Here is the history. Something happenned and we trying to explain
> why and how, and what does it mean for present if anything. Post
> factum, not before something occurs!!!

Agreed. But in history, we seek to explain an outcome (fall of Rome) by reference to a prior state. The fall of Rome was a long and complex process that in retrospect seems to have been headed in that direction. We tend to see events in terms of the extent to which they hastened or retarded that decline and eventual fall. The system as a whole acts _as if_ there were a telos, and that directionality arises from the structure of the system, and the potentials for its future development are constrained by the age of the system. No one today would suggest that the fall was the outcome of a single devastating act. Although there are examples in history where the ruling elite seems to have decided at some point to give it all up, their doing so was apparently a result of their understanding that they had no other option.

Haines Brown

Iaroslav,

Thanks for your challenging reply.

Iaroslav wrote:

> “2) As Your example of Bolivia goes, my answer still the same. It is
> wild guess, if You want with high probability of success but still
> wild guess.”

Haines Brown:

I don’t know that we are disagreeing here. You don’t seem to object my suggestion that Bolivia will probably be here tomorrow, although you label it a “guess”. I don’t much care for that word here because it implies an opinion without knowledge or means of knowledge or a random judgement. A suggestion that Bolivia will be with us tomorrow does not seem a random judgement, but one based on the knowledge that it has persisted many years, is well institutionalized, etc., and so I feel I can project its existence to some degree into the near future. You don’t seem to disagree on the substance of the point as you follow it with:

Iaroslav:

Haines:

Iaroslav:

Haines:

Perhaps I should not have spoken so abstractly, and so here's an example that is somewhat more specific:

Iaroslav:

Haines:

I guess it's time for a concrete example, drawn from a less contentious subject than historiography.

Iaroslav:

Haines:

Haines Brown

Mp> Thanks Haines,

I was sure that my posting from June 27 will be my last on this subject. But….

First. I do not know how I more could clearly stated my disagreement with Your position. In Your posting from June 9, You wrote:

“As a practical matter, prediction in history is certainly possible. For example, if a situation is relatively simple and the prediction is short range, we do it all the time. We are almost certain that next week the Republic of Bolivia will still exist.”

I was and am saying—that types of predictions ( “Existence of Republic of Bolivia”) do not have anything to do with history. It is not HISTORY.

Now You are saying:

“I don’t know that we are disagreeing here. You don’t seem to object my suggestion that Bolivia will probably be here tomorrow, although you label it a “guess”. I don’t much care for that word here because it implies an opinion without knowledge or means of knowledge or a random judgement.”

All what I was and am saying that HISTORY DOES NOT CARE ABOUT FUTURE. HISTORY DOES NOT KNOW ANYTHING ABOUT SOMETHING THAT EVEN DO NOT EXIST YET. If You could not see it, I am sorry. It is probably my fault. I could not be more clear on that.

Second, I do care about the words which we are using. About that I was clear in my previous reply. And again, if one does not care about meaning of the word which one uses than it could not be any discussion. ( See my reply to You in posting from June 16—“ empty metaphysical bias”).

Third, it is clear for me that we have totaly different views on “ What is history?”. May be I am wrong ( not first and ,I beleive, not last time), but I am pretty sure that all this differences are clear to You too.

Regards,

Iaroslav

Iaroslav,

While this thread is rather dragging on, I suspect there some rather important issues for which we seem potentially able to move toward agreement, or at least better understanding, and so it seems worthwhile.

> > As a practical matter, prediction in history is certainly
> > possible. For example, if a situation is relatively simple and the
> > prediction is short range, we do it all the time. We are almost
> > certain that next week the Republic of Bolivia will still exist.
>
> I was and am saying—that types of predictions ( “Existence of
> Republic of Bolivia”) do not have anything to do with history. It
> is not HISTORY.

Well, yes, I agree to some extent, although with one proviso. Wouldn’t you agree that in any historical situation there are elements of continuity and elements of novelty? If so, then we might be in accord in that what is of primary interest to the historian is what is novel rather than what represents continuity.

I say this to put you into an agreeable frame of mind, but I have two qualifications that you may not find so appealing.

First, the baggage we inherit from the past (representing the element of continuity) seems to constrain our actions in the present. For example, where I happen to live there are two dominant political parties that are well established (their hegemony represents an element of continuity). When it comes to voting, people often say that voting for a small third party would be a waste of my vote because third parties can’t win. Given this, the hegemony of the two main parties constrains the viability of my possible choices. That is, the element of continuity in history does seem to constrain people's behavior (determine the probability distribution of its possible outcomes and therefore the wisdom of our choices), and if so it is certainly relevant to historical explanation even if it is not an expression of the creativity that naturally attracts our attention.

Second, people today are inclined to see things in terms of systems, which represent a structural continuity that persists despite their considerable empirical change. For example, I happen to live within a capitalist system that, despite rather profound changes, nevertheless seems to have been “essentially” the same for well over a century. The existence of that system modifies the outcomes of my behavior (this used to be called “structural determinism”), making them not a simple reflection of my intentions. The Great Depression occurred ‘though no one wanted it. Here, again, an element of continuity—the structure of the system in which action takes place—must be taken into account to explain novelty. In short, systemic continuity seems necessary to include in historical explanation.

Now, I readily admit that there are elements in my two exceptions that are open to debate, but that would distract us. So let me merely say that the two points are considered valid by a significant number of people who feel that the study of historical novelty can’t be understood without taking into account elements of continuity. Any action depends on and builds upon the inheritance of the past, and so it can’t be understood without taking the past into consideration.

> All what I was and am saying that HISTORY DOES NOT CARE ABOUT
> FUTURE. HISTORY DOES NOT KNOW ANYTHING ABOUT SOMETHING THAT EVEN DO
> NOT EXIST YET.

You are correct: historians generally study the past, although historians of “contemporary history” get perilously close to the present. However, the point I tried to make was really about the method of analysis, not the object of study. I suggested (admittedly rather boldly), that the prognosticator (e.g., meteorologist) and the historian perhaps share a method that seeks to analyze emergent (tending toward outcomes that are improbable in relation to some initial state) processes in time. The truth offered by the meteorologist is only a probability distribution; the truth offered by the historian, because it is informed by the eventual outcome of a process, is an explanation in specific terms of the potentials for action in a given situation, the degree of struggle needed to achieve a desired outcome or why the intended outcome was not realized.

If this be granted for the moment, then the historian should be able to “predict” the future in the limited sense of acquiring knowledge of the probability distribution of possible near future outcomes. Admittedly, historians have been prone to failure in any such prognostication, but so too are meteorologists, which does not seem to discourage them. The reason is that the truth they offer is only probabilistic, not mechanically deterministic. But this is useful knowledge and as much a truth as an absolute prediction that it will rain tomorrow. I believe relevant here is that in the past, historians have often been called upon by states to advise them on policy. While that may not be the natural function of an historian, who prefers to use the known outcome of a process to determine the real potentials existing in some prior state (”the Owl of Minerva takes flight only at the gathering of dusk”), I see no reason in principle why historians should not enjoy a certain degree of wisdom about the potentials of their present situation. I attribute their failures to be more the result of ideological biases than incompetence.

True, some of the decisions implied here, such as avoiding unequivocal prediction, may be the result of conventions we associate with the historical profession. But if so, that would be yet another example of the constraints imposed on our behavior by continuous factors.

> Third, it is clear for me that we have totaly different views on
> “What is history?”. May be I am wrong ( not first and, I beleive,
> not last time), but I am pretty sure that all this differences are
> clear to You too.

Yes, I do get the sense that you may be seeing the study of history in terms different than my own. However, you have not specified what you mean by historical study and truth, and so I can’t offer any specific response to your observation. In lieu of that, let me just list some generalities representing my own views so that you can pin down where our disagreements may lie:

1. I adopt the axiom of scientific realism that unobservables (events, processes, causal relations, probability) are objectively real, not just mental constructs lacking truth value. This implies a rejection of empiricism (positivism, neoempiricism, logical positivism, etc.) Being an axiom, it is not debatable, although it should be justifiable.

2. I believe the historian can choose to study social processes in time. By “social” I mean the object of study is holistic, taking into account such things as the system's environment, economy, geography, and individual choice (a la the Annales School). I would argue that to the extent a subsystem under study is isolated from its environment (the larger systems of which it is part), we cannot account for human creativity without an implicit appeal to supernatural forces.

3. Individual behavior necessarily arises from and contributes to these social processes, but it is not the center of my attention. The reason is that I do not believe individual behavior can be explained in isolation from its social environment (for the reasons outlined above). I believe that the goal is to understand human behavior in terms of the situation in which it takes place, that we are creative because we are social beings.

4. Further, I believe that this “situation” that we inherit from the past, if understood as an open system, entails a combination of limits and opportunities—of constraints and potentials. Given that the system is fuly open, these constraints and potentials define a probability distribution of possible outcomes, and human actions must be assessed in terms of them.

5. Finally, by applying this historical method to our present situation, we better understand the possibilities and likely outcomes of our behavior and the struggle required to achieve our ends. This is why “historical consciousness is a cornerstone of human freedom” (Lord Action, roughly).

Let me emphasize that I’m not suggesting this represents the only valid kind of historiography. Narrative history, for example, has its value and legitimacy, as does art or economic history. The question is then, why do I prefer the kind of historical study outlined above? First because I believe the study of subsystems taken in isolation yields only one-sided knowledge, although even one-sided knowledge has some use and interest, and it seems a precondition for achieving a more universal understanding. Second, it is because I believe it best supports wise choices in the present. However, not everyone is as concerned as I am about the relation of historical understanding to what is happening in the world today.

Haines Brown

Haines, hi

With all my disagreements with Your point of view, I do not see any misunderstanding of our positions.

1) Novelty and continuity.

Existense of continuity and novelty in any given situation is trivial. But for any person ( and , of coarse, historian) significance of continuity or novely depends on personal interests very often. Actually for me, more puzzling is continuity.

To show what I mean, I give You an example of Russian empire/ USSR/ Russian Federation. There are significant and very important differencies between all of them, but when one starts to look in history of Russia one will see undoubtetly prevalence of strong state above civic society( in western meaning of civic society). Why is that? Are the any clues in history, which could help me understand why after collapse of Communist party and great liberalization of Russia in 90, now under Putin old sturctures appear to coming back? Of coarse, those old sturctures of goverment and society are changed. But novelty in them is not so great for me not to see continuity of traditional Russian of interaction in between state and society). You could say that I in agreement with You at that point. But actually for me it is not. Mine or somebody else predictions what could happened after next presedential election in Russia, does not have anything to do with history. One does not need to know history of Ivan the Terrible and Time of troubles, or history of Bolshevicks, or Stalin and time of troubles after his death ( Beria-Khrushev-Malenkov struggle for power) to try to predict results. Much more succesful prediction of successor to Putin (if there will be one, I mean that this guy could always try to change rules of the game. Nothing new from Russian history, but he(Putin) may decide not to change rules of the game. Who knows!) could be done based on analisys of recent development in Russia ( political, economical,demographical and so on). And of coarse, You are right when You are saying that better predictions could be made based on understanding of history, but it does not mean that prediction of development of political situation in Russia is history. History even as organized body of knowledge ( that was one of defeniton of science which You gave in one of Your postings) does not know what could happenned. It (history) could not have knowledge of something which does not exist yet.

2) Another very importent question which You asked is what I mean by “historical study and truth”.

About truth that's actually easy. And You know answer as well. Is there answer on this biblical question “ What is truth?”. But Peter Watson in his “ Ideas: A History of Thought and Invention, from Fire to Freud “ quotes Whithead on truth as epigraph. That words will be pretty close to my defenition of truth. ( Unfortunately, I do not have this book with me at this time. Otherwise I will quote those words).

In regards to “historical study”. Original question of disscusion was about is history science or not. I beleive I was clear enough on this subject in my posting from June 5, 2006. As far as I could see my position is more or less covered by definitons a) and c) which were provided by You in Your posting from June 19, 2006. The problem there as I wrote in other places and as You could understand, that in contemporary mind word science is relatevily heavily connected with phisics and mathematics. Again, if one looks carefully in all methodology of science, he/she will find out that scienticity of discipline is heavily judged by usage of mathematical/physical apparatus or system of agumentation. For example, modern study of complex systems were originated from phsysics ( one of the first examples was usage first prototypes of computers in calculations of hydrogen bomb for Edward Teller project). Later on this system analyses were applied to another disciplines. But the problem there for me personally, that all this system analyses does not work very well with really complex systems. Whenever system analyses has to deal with highly complex system, it has to simplify situation. I do not care, figuratevely speaking, abbout atoms and protons, but I do not think that simplification of human society for the purpose of showing how good one's model work is proper way to go. I do not beleive that application of any simplified model, even “fully-open system”, will significantly improve our understanding of past and present. That's why I prefere not to use “science” as adjective to history.

Regards,

Iaroslav.

Iaroslav,

Glad to see you took this discussion off list. I was feeling rather guilty.

> 1) Novelty and continuity.
>
> Existense of continuity and novelty in any given situation is
> trivial. But for any person ( and , of coarse, historian)
> significance of continuity or novely depends on personal interests
> very often. Actually for me, more puzzling is continuity.

I agree to some extent that whether it is continuity (as in your example of Russia) that is curious/interesting/puzzling or whether it is novelty (such as the Russian Revolution) depends to a significant degree on the observer's personal inclinations. I can think of many examples where this is so.

Personally, however, I’m a bit uncomfortable just reducing it to a question of personal tastes. There are two reasons for this that I’ll not elaborate unless you press the issue. One is that I have reason to take novelty for granted, and so it becomes continuity that begs for explanation. This may be in part because of my interest in world history, for the larger the scope of our concern, the more will change and complexity be apparent; the narrower our focus, the more will uniformity and continuity prevail. The second reason is ideology. We have ideological reasons to favor one or the other. To put it very crudely (coming close to reductionism), those in positions of advantage prefer continuity, while those who are not look for change. This bias tends to make one aspect or the other more interesting and worthy of study. (If you think you perceive a contradiction in this paragraph, you should know that I believe it can be resolved.)

While we personally, because of our interests and situation, may focus on one or the other aspect of history, nevertheless the historian is in the business of communicating, and so his or her predilections beg for justification.

> Mine or somebody else predictions what could happened after next
> presedential election in Russia, does not have anything to do with
> history.

You may have missed a point I made in my last message. I was speaking of the _method_ appropriate for the study of emergent processes, not simple prediction. If anticipating the consequences of the Russian election is meant as an example of actual prediction, I’d quite agree with you. But analysts of the outcome of the election (not specifically historians), do in fact say something like this: given the electoral result, it could well have such and such a consequence. What is stated here is not a prediction, but the assignment of a probability to possible outcomes. It is a different kind of knowledge.

Now whether this has anything to do with history depends on what we mean by “history”, If we take it to mean the historical profession or the practice of historiography, you are right, for most historians study processes the outcomes of which are known. But my suggestion was that the method employed by the analyst of the election and the historian analyzing some past situation is much the same. This is because the explanation of emergent processes entails an entirely different method than the hypothetico-deductive—one of abduction. The only difference is that the analyst of the election does not know what will happen in the future, and so can only offer a probability distribution for possible outcomes. This is what the meteorologist does. The historian, on the other hand, does know the outcome. As a result, he is in a position to investigate what caused the actual outcome to take place in relation to the probability of various possible outcomes.

This gets into a very interesting problem. The past of an historical situation under study tells us the probability distribution of its possible outcomes; the future of the situation (the known future outcome) tells us which of those possibilities was in fact realized. But in human history, we generally are concerned with emergent systems, and so human intentions constrain the probability distribution to yield a relatively improbable outcome, not what was most likely. I leave aside how an idea (a dream, a goal, a plan), which have no real potency in themselves, can actually bend the course of history. The point is that there's no prediction in any of this. When we act, we hopefully are aware of its likely consequences, this is not a prediction, but only stating the probability distribution of the possible outcomes of our action.

> History even as organized body of knowledge ( that was one of
> defeniton of science which You gave in one of Your postings) does
> not know what could happenned. It (history) could not have
> knowledge of something which does not exist yet.

That definition of science is just a very conventional one and I only mentioned it to point to one feature that natural science and historiography have in common (there are some historians, as you know, that do not see history as an organized body of knowledge, but that's probably irrelevant here). The real division, I believe, is between those sciences studying dissipating systems and emergent systems (which are sometimes called “dissipative systems” because emergence [or the maintenance of far-from-equilibrium systems] require the engine of dissipation), which can be confusing), for the methods employed are quite different.

> 2) Another very importent question which You asked is what I mean by
> “historical study and truth”.
>
> About truth that's actually easy. And You know answer as well. Is
> there answer on this biblical question “ What is truth?”. But Peter > Watson in his “ Ideas: A History of Thought and Invention, from Fire
> to Freud “ quotes Whithead on truth as epigraph. That words will be
> pretty close to my defenition of truth. ( Unfortunately, I do not
> have this book with me at this time. Otherwise I will quote those
> words).

It would have been helpful. I could just as well suggest that truth is successful action. “Truth” is a tricky question in philosophical terms, but I meant the word in a simple sense that the truth value of a statement is the extent to which it represents reality in thought. That is, Rome fell for a reason, even if I don’t happen to know it or historians can’t agree upon it; my explanation of Rome's fall are arguments (words) that have some relation with the real reason, whether it have much truth value or not. Yes, some would contest this, but I have no reason to believe you are one of them.

> The problem there as I wrote in other places and as You could
> understand, that in contemporary mind word science is relatevily
> heavily connected with phisics and mathematics.

That is true, but only because the “contemporary mind” has no idea what the scientist is up to. I happen to believe historians have more to gain by associating themselves with science as it is actually practiced than dissociating themselves from a popular misunderstanding of science. I say this not in terms of whether the historians can “sell his product”, but because I believe the historian would have much to gain from the rigor and discipline expected in the sciences, as well as some understanding how all things must work.

> Again, if one looks carefully in all methodology of science, he/she
> will find out that scienticity of discipline is heavily judged by
> usage of mathematical/physical apparatus or system of agumentation.

I must disagree with this as a generalization. Physical apparatus (discounting the observational apparatus, which the historian also uses), applies to only one kind of science. Cosmology is a science, and is not (for the most part) subject to laboratory experiment. Some approaches to evolutionary biology are highly mathematicized, but some are not. The sciences are not simply different expressions of mechanics.

> For example, modern study of complex systems were originated from
> phsysics ( one of the first examples was usage first prototypes of
> computers in calculations of hydrogen bomb for Edward Teller
> project). Later on this system analyses were applied to another
> disciplines.

We may differ here. The computer has much deeper roots than atomic bomb studies (a programming language, Ada is named after the woman [Lord Byron's daughter!] who is widely considered the pioneer in conceiving of a computer), but were computers not used more to study system complexity than complex systems? And that seems still the case. That is, a system with many variables rather than how complex systems behave as a result of their complexity. General systems theory did start during World War II (Norbert Wiener's Cybernetics—and, yes, he was involved in atomic bomb development), and blossomed after the war with people like Blauberg, Sadovsky and Yudin (Systems Theory [Moscow, 1977]) and Herbert Simon (The Sciences of the Artificial [Cambridge, 1969]). General systems theory and computers represent quite different concerns and are only marginally connected. The latter are mathematized; the former generally not.

> But the problem there for me personally, that all this system
> analyses does not work very well with really complex systems.

Again, I suspect you have in mind mathematical model-building here, in which case I fully agree.

> but I do not think that simplification of human society for the
> purpose of showing how good one's model work is proper way to go.

One of the features of evolutionary sciences (which study emergent systems) is that they start with complexity and do not seek to explain that complexity, but the conditions in which unique and complex outcomes might emerge. I hope I was not hinting of any fondness for simplified models.

> I do not beleive that application of any simplified model, even
> “fully-open system”, will significantly improve our understanding of
> past and present.

Fully open systems do not in themselves account for emergence (some open systems dissipate), but they certainly do not represent a simplification. Far from it. The reason is that a fully open system has virtually infinite determinations, and you can’t even begin to account for all of them. Simplification, mechanical determinism (and the possiblity of mathematicization), are associated only with artifically closed or isolated systems (there is an exception here - thy physics of open systems, but it is esoteric).

I believe the only place I got close to a simplification was when I discussed the structure of a system such as capitalism. That is, a capitalist economy “hangs together” in such a way that one gets a sense of its dynamic. However, this only refers to its causal relations and does not at all entail empirical simplification. The word “model” is highly ambivalent, and so I hesitate to comment on it.

By the way, this is my objection to Gundar Frank, etc., “world systems theory”. It isolates one aspect of life to discover in it regular behavioral patterns that are in part an effect of that isolation, and from this concludes that the object of study is a “system”, and then it uses that system as a basis for an explanation of other aspects of life. I believe this is reductionist and circular.

> That's why I prefere not to use “science” as adjective to history.

Given your notion of science, I can understand why you come to this conclusion, and we are not really disagreeing. However, I’m also suggesting that there's a host of “evolutionary sciences” that use a method that is the same as that of the historian, and if that is the case, there are good reasons for historians to embrace the label “science”.

Haines

Haines,

Your message is too big, as was mine as well, and I’m too lazy to quote parts on which I am replying. As I could see, Your e-mail consists of 4 main parts. I beleive You have copy of Your e-mail so I will just refering to parts as I see it.

1) “Continuiity” and “novelty” in history and historians.

Of coarse I may be oversimplified my point but what justification one may ask of Einstein interest in relativity? He ( Einstein) did probably have some explanations, but reality is much simplier. All humans are different and different persons do have different interests. So, even if I could what difference does it to me to know why Braudel did study mostly era in between 15-18 century. What is important to me he ( Braudel) did produce very intersting and usefull works about this period! And this result justified his job in itself. On other side, some soviet historians produced volumes of justified usleless non-sense. But again, their ( those historians) justification does not change the simple fact that their work is, I am sorry to say, garbage.

2) “Method appropriate for the study of emergent processes”.

Here we are going again. Are we discussing “what is history” or what methods could be used in this or that investigation? All what I was saying in quote which You are using that those type of predictions ( e.g. “ Republic of Bolivia” or “presidential elctions in Russia”) are not history. Your suggestions of analysis of possible outcomes of the future elections as assignment of a probabilty…” is not history what so ever. It is futurology. It is just totaly different discipline from my point of view.

But below You are coming back with “…what we mean by “ history”. So what are we discussing : “the method appropriate for study of emergent processes” or “ what we mean by history”? If You are discussing or want to discuss methods appropriate for use in this or that discipline, than let discuss methods and their approprietness for use here or there. If You want to discuss “what is history”, than let discuss it. This is two different subjects. The fact that archeology is using radio-carbon analysis does not make archeology part of physiscs. So the fact that knowledge of hsitorical facts could help in analysys of electroal distribution and prognosys of results, does not mean that it is history. This analysis and forecast is futurology and bussines of futurologist. This person/ futurologist may be or may be not historian. It does not matter. What I am saying all this discussion about probabilty and possibilty of this or that ourcome is not history. All what You are doing is using history similiar how archeologist could use genetics to study to analyse and speculate ( with higher or smaller success) predecessors of Russians people in Eastern Europe. But it is still usage of history, not history itself!!!

3) Dissipating and emergent systems.

You already know my response to that ( see my posting from June 16, 2006).

4) Truth I do not have anything to add to my previous e-mail.

5) Science.

Cosmology is a part of phisics ( I am not talking of methaphysical or religious cosmology, I do not see how their subject could be related to this discussion at all). May be You have different understanding of what is it but that's how I see it. About evolutionary biology, I could not tell You I am very bad at biology in general ( beyond bad). But as long as You mentioned evolutionary biology and system analysis, could I suggest for You two books of P. Turchin “Historical dynamics” and “War and peace and war”. Turchin is evolutionary biologist and mathematician. Based on his success in ecology/ evolutionary biology, he tried to apply his modelling of emerging processes to hsitory ( based on system analysis). “Historical dynamics…” is theory. “War and peace and war” is attempt to application. Sorry it does not work for me. It is just another unsuccsesful attempt of usage application of math and system analysis to historical facts. It does not make history.

6) Computers, cybernetics and so on.

If You want to discuss roots of computer, You could start not from daughter of lord Byron but from “abacus”. Question is how deep You wanna go.

You given me name Norbert Winer. Who was he? Was not he first mathematician and after creator of cybernatics? Was not his interest in mathematical logic influential in creating of cybernatics?

Yes, creator of general system theory Ludvig von Bertalanffy was biologist. But did not he addmited that his general system theory is mathematisation of Hartmann's ontology ? ( see wikipedia.com on Bertalanffy).

I was not mean to say that You simplified something ( If that's how You understood me that mean I was not clear). All what I am saying that human society is too close to me to use simplification which is unavoidable when one has to use approaches of math ( even in disguise of system analysis). We are not digits and numbers, we are humans. Our human logic or luck of logic could not be described by method or apparatus of mathemathical logic/ system analisys.

6) Conclusion.

As I stated several time before, it is pretty clear to me ( and I beleive to You too) where each of us stands. There are some common grounds, but differences are much bigger. I do not see any problem with that. Again different people—different visions.

Iaroslav

Thank you for your patience with my long post. It, of course, was in part an exercise in self-clarification, and so I indulged myself. > 1) “Continuiity” and “novelty” in history and historians.
>
> Of coarse I may be oversimplified my point but what justification
> one may ask of Einstein interest in relativity? He ( Einstein) did
> probably have some explanations, but reality is much simplier. All
> humans are different and different persons do have different
> interests. So, even if I could what difference does it to me to
> know why Braudel did study mostly era in between 15-18 century.
> What is important to me he ( Braudel) did produce very intersting
> and usefull works about this period! And this result justified his
> job in itself. On other side, some soviet historians produced
> volumes of justified usleless non-sense. But again, their ( those
> historians) justification does not change the simple fact that their
> work is, I am sorry to say, garbage.

There seem three issues here: a) the intrinsic quality of a piece of historical research, b) The manner in which that research reflects the interests of the writer, c) My point that what we do requires justification, which you apparently deny. I suspect your inclination is to go from (a) to (b) and stop short at (c); mine is to start at (c) and then move from (a) to (b). No point in getting off track and arguing about this, but do I detect a contradiction in your point about Braudel? If his work is “useful”, does that not imply its justification? Perhaps you don’t object to (c) after all, but I promised I would not belabor our difference here.

> 2) “Method appropriate for the study of emergent processes”.
>
> Here we are going again. Are we discussing “what is history” or
> what methods could be used in this or that investigation?

Both. I assume that different objects of study potentially have different methods necessary to comprehend them. I believe you agree with this point. My point simply: “emergent processes” is a generally well understood kind of process; the scientific study of emergent processes requires a particular methodology; since history is an emergent process, I suggested that science may have something to offer the historian by offering an explanation of the conditions necessary for an emergent process.

> All what I was saying in quote which You are using that those type
> of predictions ( e.g. “ Republic of Bolivia” or “presidential
> elctions in Russia”) are not history. Your suggestions of analysis
> of possible outcomes of the future elections as assignment of a
> probabilty…” is not history what so ever.

I agree, it is not history. Virtually by definition a present event is about the present, not the past, and so is not history (no “story” of what has happened). But, again, my point was about _method_ appropriate for the study of emergent processes, and some emergent processes are taking place right now, and some took place in the past. You are correct, figuring out what the outcome of an action in the present is likely to be is futurology and has nothing to do with history. However, if I am able to arrive at the probability distribution of the possible outcomes of some action in the present, it is _not_ futurology; it is a statement about the present, not about the future.

Historians employ the abductive method, only they (in the English-speaking world, anyway) generally call it “retrodiction”. In deduction, you start with an initial state to explain an outcome; in abduction you start with an outcome to reveal a truth about the initial state.

> But below You are coming back with “…what we mean by “ history”.
> So what are we discussing : “the method appropriate for study of
> emergent processes” or “ what we mean by history”? If You are
> discussing or want to discuss methods appropriate for use in this or
> that discipline, than let discuss methods and their approprietness
> for use here or there. If You want to discuss “what is history”,
> than let discuss it.

Granted, these are two different questions, but I thought I had made clear early on that I was talking about was method. I avoided the question, “what is history?” because, frankly, I have no answer. In a very broad sense, it is the story of the past, but there's no story of the past that does not include interpretation in some fashion. A simple chronicle of events entails a culling of significant facts from what is trivial, and significance is in the eye of the beholder. So history is not just the story of the past, but our interpretation of that past. That interpretation is necessarily ideological, and so history is not only about the past, but about the present, etc., etc. It gets complicated.

> This analysis and forecast is futurology and bussines of
> futurologist. This person/ futurologist may be or may be not
> historian. It does not matter.

That actually helps make your position clearer, and it is not one with which I would disagree (with the exception of the point above about establishing the probability distribution of possible outcomes, which I don’t believe to be futurology). Simply put, the historian acting as a futurologist is not acting as an historian.

> What I am saying all this discussion about probabilty and possibilty
> of this or that ourcome is not history.

I fear you may be overly bold here. The language employed by historians habitually employs the language of probability and possibility. Of course that language is not applied to the future, but to a past process. Surely there's nothing intrinsically wrong with speaking in probabilistic terms.

> All what You are doing is using history similiar how archeologist
> could use genetics to study to analyse and speculate ( with higher
> or smaller success) predecessors of Russians people in Eastern
> Europe. But it is still usage of history, not history itself!!!

Not sure I follow you here. I seems you are referring to what (in English-speaking historical circles) is called “auxiliary sciences”. They useful expose the facts from which history is constructed. This in itself does not make history a “science”. Such sciences are merely tools, not the essence of historiography. I suspect we agree here, but why I’m having trouble following you is that I don’t understand your point. If it is that history is a distinct discipline, having its own specific character that distinguishes it from, say, genetic studies, then I fully agree. However, the same could be said of the difference between genetic studies and thermodynamics, and so merely that historiography is distinctive does not in itself demonstrate that historiography is not a “science”.

Some argue that history is a science in that it is an organized body of knowledge and benefits from the application of the intellectual rigor and honesty we associate with proofs in the natural sciences. This in itself seems a harmless point, and the real issue seem instead to be all the other garbage that the word “science” might imply and which seems alien to what the historian is up to. What I’ve been trying to do is to suggest that the common association of this “garbage” with science is often not warranted.

> 3) Dissipating and emergent systems.
>
> You already know my response to that ( see my posting from June 16,
> 2006).
>
> 4) Truth I do not have anything to add to my previous e-mail.
>
> 5) Science.
>
> Cosmology is a part of phisics ( I am not talking of methaphysical
> or religious cosmology, I do not see how their subject could be
> related to this discussion at all). May be You have different
> understanding of what is it but that's how I see it.

You lost me here. In what sense is cosmology part of physics? I pluck a classic physics textbook off the shelf, and while it has chapters on relativity and quantum mechanics, it says nothing about cosmology.

There is an argument that all phenomena (everything from the Big Bang to a poem), exist in a hierarchy, in which each “level” (another difficult term) emerged from a prior level, ultimately going back to the big bang. Each level is a specification or constraint of the level from which it emerged and remains subject to its laws. Thus human society is subject to the laws of physics, although does not reduce to them, for society emerged and acquired behaviors peculiar to itself.

> About evolutionary biology,

I’ve lost the context here. Not sure at this point why I mentioned evolutionary biology. As for Turchin, I took a look at a synopsis of his “Historical Dynamics”, and would probably agree with you: it seems reductionist, and I’m not enthusiastic about empirical model building (I have somewhat the same objection to Braudel). What seems missing in such approaches (well, a lot seems missing), is that territorial dynamics seems to leave the human element out of the equation. He is using a system analysis, but in a rather specific way with which I would disagree, for it seems basically functionalist, which if true leaves no room for emergent systems, however dynamic they may be.

> You given me name Norbert Winer. Who was he? Was not he first
> mathematician and after creator of cybernatics? Was not his
> interest in mathematical logic influential in creating of
> cybernatics?

If your point is that he was a mathematician? Yes? But so? He was also a character! I remember him running his finger along the wall counting doorways so that he would enter the correct classroom. Again, I’ve lost the context of this discussion of computers.

> Yes, creator of general system theory Ludvig von Bertalanffy was
> biologist. But did not he addmited that his general system theory
> is mathematisation of Hartmann's ontology ? ( see wikipedia.com on
> Bertalanffy).

I never cared much for Bertalanffy, which is why I cited the book by Blaubert et al. Let me quote them briefly:

…the concept of ‘general system theory’ appeared in Bertalanffy's works as a generalization of the principles of the ‘the theory of open systems'. Open-system models, and the tools used for their evolvement, covered—already at the time when the theory was created—not only biological phenomena, but also facts from chemistry, physical chemistry, psychology, sociology, etc.

Human history unquestionably has to do with open systems, although not all open systems are emergent, just probabilistic. Further, open system defy measurement (mathematicization).

Well, enough of this ;-)

Haines

Iaroslav,

Haines:

Iaroslav:

“I was and am saying—that types of predictions ( “Existence of Republic of Bolivia”) do not have anything to do with history. It is not HISTORY.”

I say this to put you into an agreeable frame of mind, but I have two qualifications that you may not find so appealing.

Iaroslav:

“All what I was and am saying that HISTORY DOES NOT CARE ABOUT FUTURE. HISTORY DOES NOT KNOW ANYTHING ABOUT SOMETHING THAT EVEN DO NOT EXIST YET.”

Iaroslav:

“Third, it is clear for me that we have totaly different views on “What is history?”. May be I am wrong ( not first and, I beleive, not last time), but I am pretty sure that all this differences are clear to You too.”

Haines Brown

This thread spun off into a very pleasant and useful private discussion with Iaroslav off list. One result of it has been a clarification of a misunderstanding that seems important to offer here as a possible resolution of the issue of prediction, which was left in suspension in this thread.

If we are able to assess in a historical situation under study the probability distribution of its possible outcomes (something which, arguably, all historians do), then this does not entail any prediction. At the risk of speaking for him, my impression is that Iaroslav implicitly agrees with this.

A prediction has truth value in relation to the future of the situation under study. On the other hand, an assessment of the probability distribution of its possible outcomes has truth value only in relation to the actual situation being studied. It is an assessment of its potentials for change and the structural constraints upon them.

There may be involved here a presumption of a realist philosophy that holds unobservables (events, processes, causal relations, in contrast with facts subject to empirical observation) to be objectively real objects of study and not merely mental inferences about which we cannot really assess a truth value. I’m interested in people's thoughts about this.

One objection to the presumed pernicious role of theory in historiography (what Roger Collins once called a “German vice”), is that it builds castles in the air out of our concepts, not facts. And yet, if our “facts” included real unobservables, then theory would seem to offer a “realistic” tool for understanding the past.

Haines Brown