Revising historical accounts when new evidence comes to light or when old evidence begs to be reconsidered is surely one of the major tasks of the scholarly historian. The practitioner should ideally possess the ability to unravel the past before our eyes in a clear and unbiased fashion, both at phenomenological and causal levels. Some would argue that a scholarly historian can not, any more than can a microbiologist investigating the minuscule parts of an organism, detach the circumstances of his or her act of observation from the object itself. In this view, historians' accounts of the past are subjective narratives by default if not by design. Narratives can, it might be argued, be disproved—but factuality cannot be proved. Yet even this is not as straightforward as it might seem: the best one can do is simply to accumulate narratives, not with a view to displacing older with newer ones, but to develop a larger or, pointedly in this context, more useful, picture of reality.

By choosing two major events in Israel's history (the 1956 war and the Six-Day War), and corresponding but conflicting (Israeli) historians' accounts of them, Jonathan Isacoff presents a paradigm of the historicist's (or political scientist's) dilemma and seems to offer a “soft” method for dealing with it. He views narratives simply as such; that is, they are perspective-related, rather than independent and neutral accounts of an objective reality. With this soft approach, (new) narratives need not be considered as absolutist truth-bearing accounts, and they should not therefore be viewed as competing for a reality-legitimacy with, or seeking to displace, older accounts. Old and new accounts alike can be considered perspective-legitimate, rather than objective-legitimate. Two presumed accounts that are inconsistent with each other can thus be viewed as legitimate narratives; the political scientist's selection of one or the other can be measured by an extraneous or additional standard.

On the other side of the spectrum, one can surmise that an opposite, say “hard,” approach would not brook such a relativistic conclusion. The hard approach assumes that a historical account's legitimacy can only be judged against reality, that is, as a function of the reality it purports to describe (surface level) and to explain (causal, or deeper level). Thus it is no more possible to entertain two inconsistent accounts of a historical event than it is to accept two inconsistent (or even contradictory) descriptions of an object. It stands to reason that only one of the two accounts is true, while the other is false, or that they are both false. Both cannot be true. Thus, for example, either Israel planned to execute the Six-Day War with a view to expansion, or it did not plan to execute that war and was compelled to enter it against its will. (The implicit claim that Israel both planned the war and did not plan it is meaningless.)

Isacoff's thesis, however, does not ostensibly, or necessarily, seem to lend itself to this criticism; indeed, instead of allowing himself to be stuck by the classical but “futile” dichotomy of proposing either an objectivist or a relativist approach, Isacoff opts for a pragmatist-inspired third approach that could be described as “heuristic”: an account is to be judged or valued (by a political scientist) against the fulfillment or otherwise of what might be called two extraneous—that is, veracity-independent—conditions, to which I shall return below. For the moment, these two conditions can be summarized as having to do with comprehensiveness and consistency on the one hand (providing internally consistent answers to any and all relevant questions to a problem), and instrumentality on the other (contributing to a resolution for that problem). However, it is not readily obvious how or whether a pragmatic approach might escape the usual objections to relativism. As to the noncontradiction criticism, this is, in any case, circular in that it presupposes absolute objectivity, which is precisely what the soft approach sets out to cast doubt on. Furthermore, we do not have to read Isacoff's heuristic position as being predicated on rejecting the possibility of an absolutist account, but only as one that, in the practical absence of such an account, provides political scientists with a way to use conflicting accounts by pointing out a reason for pursuing a wider spectrum of narratives while provisionally abstaining from imputing a truth-content to any one of them that outweighs or cancels the truth-content of another.

However, such an interpretation, while provisionally sparing a heuristic approach the criticism of being too soft, is only a stopgap measure: eventually one who defends this approach must determine whether or not there is a truth “out there”—and a corresponding account. To hypothesize that no such truth exists means having to redefine the historian's project: his or her task becomes not to describe and explain objective reality, but rather to collect perspectives or narratives and perhaps to choose (according to some set of criteria) the “best” from among them. But this way of dealing with mutually inconsistent accounts (e.g., the Zionist and the new Israeli historians' accounts), raises the fundamental question underlying Isacoff 's thesis: can (or should) a history be treated as one might a mathematical or scientific model? Here we need to expand on the “extraneousness” of the two Isacoff conditions. In the absence of the possibility of an objective veracity standard, it is still possible to select a narrative that satisfies the condition that it generate answers to any and all questions raised (completeness requirement) in a way that all propositions generated by the narrative do not contradict one another (consistency requirement).

Thus, if the Zionist account of the 1956 war did not incorporate any references to new evidence about the dominant internal Israeli school of thought favoring a military over a diplomatic option or any references to the institutionalization of a self-perpetuating militaristic clique, then, in light of newly released archival documents, it would not satisfy the test of comprehensiveness or consistency (given the new data, one presumes the account can no longer provide an explanation for it that does not contradict the explanations it provides for the rest of the data it incorporates). If the new historians' account addressed these subjects in addition to the issues included in the Zionist account, thus providing a comprehensive but also an internally consistent picture of what brought about the 1956 war, then it would satisfy the only necessary tests needed for adoption, namely, completeness and consistency. Veracity in this context could then be considered a useless luxury.

While Isacoff does not articulate the consistency component of this condition for favoring one model over another, it is nonetheless present when, in seeking to consider evidence that is treated by an account, there is a reference to the failure of a narrative that cannot account for inconsistent pieces of evidence (e.g., a prewar Davidian portrayal of Israel next to a postwar Goliathic reality). In any case, the consistency component is an essential part of any logical model (scientific or mathematical) that seeks legitimacy on the basis of pragmatism and outside of the domain of veracity: explanations should be forthcoming for all pieces of evidence, and these explanations must be consistent with one another. However, now we should ask ourselves whether inconsistency between two separate accounts (e.g., Zionist and new) is a sufficient criterion for adopting a pragmatic approach in the first place—leading us, that is, to drop truth in favor of utility. Surely it could be argued, in favor of an objectivist standpoint, that in providing their comprehensive and internally consistent accounts, the new historians are actually revealing many outright falsehoods and suppressed truths in the traditional accounts. In any case, it must be said that, unlike a theory about the universe, or numbers, where the ordinary person might feel neutral concerning the different explanatory models being presented, a historical account, dealing with directly experienced human affairs, leaves little room for truth-indifference. Surely (to take a present-day example), either Prime Minister Ariel Sharon wishes to dodge the road map through his unilateral Gaza option (as his spokesperson has recently unwisely revealed), or he is planning to implement the latter as part of his commitment to the former (as he promised President Bush). Ordinary Israelis and Palestinians (also, one hopes, the American administration) can not feel indifferent as to which of these two possibilities is true. Or surely, again, returning to 1948, either Rabin received his orders to evict the residents of Ramleh (as he later owned up), or they simply took flight by themselves (the Zionist account). It is somewhat “disconnected” to claim that truth here does not matter.

It is important to note that, in selecting his paradigm from instances within a relatively fixed set of parameters (Israeli historians), Isacoff makes a conscious effort to present us with a least controversial tool with which to make his argument. His effort is possible because new Israeli historians have begun to question older Israeli narratives about events in their country's past. Had he picked historians from opposite sides of the fence, his argument might have lacked the initial boost of credibility it needed. Pitting an Arab historian's view of either war against an Israeli viewpoint, while perhaps raising similar questions logically (though in an entirely different way), would probably have failed to arouse the historicist's interest. After all, it is only natural to expect inconsistent accounts under such circumstances, and even more natural to assume that only one of them is right (the other being an example of bad history).

Nonetheless, the underlying epistemic questions and principles in Isacoff's examples are the same: Is there one truth and only one possible account of it, or is there nothing out there but a set of (possibly inconsistent) different narratives, reflecting different perspectives or contexts? The new Israeli historians whom Isacoff quotes posit their accounts not simply as new perspectives, but as challenging the earlier accounts and seeking to replace them. These new historians, one presumes, are commendably motivated by the scholarly (but perhaps also political/moral) urge to demythologize the past—not by a quest to reinforce its prerevisionist epistemic edifice. (An Arab should be envious that such courageous challenges to preexisting myths do not have their counterparts in Arab historical literature.) This is precisely why Isacoff's use of them as examples is so powerful, but one suspects that this power totally evaporates once one decides to view them simply as a quantitative addendum to this epistemic edifice, or as deriving their legitimacy from their completeness and internal consistency only, rather than also from their veracity.

This leads us to consider Isacoff's second condition for evaluating a narrative, namely, its functionality with regard to resolving the problem under review: the idea here is that an Israeli decision maker captivated by the Zionist narrative (e.g., Sharon or Benyamin Netanyahu) would in all likelihood continue to make decisions informed by that narrative, thereby reinforcing the continued state of conflict being written about by historians. Conversely, an Israeli decision maker informed by the narrative of the new historians (e.g., Ehud Barak or Shimon Peres) will more likely take steps to end that conflict (e.g., Barak's decision to pull out of Lebanon within the space of 24 hours). Thus, in addition to the “scientific” requirement of completeness and consistency, Isacoff proposes a political requirement for favoring one narrative over another, namely, that it foster a worldview that will be reflected in the decision makers most likely to take the steps necessary to solve the problem under review.

Although Isacoff's observation in this context is probably true, and a historian therefore can or does indeed play a role in politics, even only as a historian, there are two major difficulties (one of which may be insuperable) with identifying this as a requirement, or a condition, for favoring a narrative along heuristic lines. The first difficulty has to do with categories. In science, one theory may indeed be favored over another to the extent that it can provide solutions to unfolding or new problems (e.g., evidence of light emissions bouncing off a black hole). However, to apply this problem-solving characteristic of a scientific theory to a historical account seems almost like requiring your choice of a new car be made on the basis of whether it will settle a future conflict you might have with your partner. The requirement is incongruous with the function. Cars are not conceived as conflict solvers, and historical accounts of a conflict are not supposed to help solve that conflict. Furthermore, to extend this argument, most historical accounts are not about ongoing conflicts. To make functionality one of two necessary requirements or conditions for its selection thus immediately makes the method useless for the majority of historical accounts.

This is one hurdle, but it is not insuperable. With regard to the question about what accounts are supposed to do, Isacoff could argue (and should indeed be understood as arguing) that the criticism begs the question, in that the critic's starting assumption is a definition for historical accounts that Isacoff suggests be replaced by a new definition. It would simply be tautologous to claim that his proposed definition is meaningless by virtue of the fact that it is different from the definition of the critic. The second part of this hurdle, however, is more difficult to overcome, unless of course one modifies what one means by this being one of the selection criteria: if this criterion is to be regarded as necessary, then the scope of historical accounts to which the selection method is to be applied becomes immediately restricted to accounts of those conflicts that are still ongoing, since only those, one assumes, might lend themselves to being affected positively (or negatively) by those accounts. For a political scientist whose focus is the present and the future, and whose study of the past is undertaken with only that focus in mind, such a restriction might not seem like a high price to pay. But for historians, historicists, and the public in general, such an approach might indicate almost obsessive “disconnectedness” with reality. Fortunately, one is not compelled to interpret the functionality requirement in this way. It could be presented, for example, as a sufficient rather than a necessary condition, whose nonapplicability to accounts of finished or forgotten conflicts does not undermine the applicability of the heuristic approach to ongoing conflicts in order to favor one account over another.

The second hurdle facing the identification of functionality as a requirement may be more serious: let us return to the two narratives at hand (Zionist and new), but assume that each of these narratives manages to present an internally coherent and consistent model. The Zionist narrative (in a brushed-up form) could claim, for example, that whether true or not, David Ben-Gurion believed the country to be under threat, and that the only way to deal with the Arabs was by making sure that Israel defeat them. Therefore, though indeed planning and consciously planting the seeds for a war with the Arab countries with a view to smashing their armies and grabbing more land from them, nonetheless what was uppermost in the Israeli leaders' minds and informed their decisions was their sense that Israel's very existence was threatened and could only be protected in this manner. (A parallel can be drawn to present-day events, where what is portrayed or marketed as a Palestinian intifada, requiring a strong Israeli reaction, is in reality a carefully preplanned and staged Israeli destruction of the possibility of a viable Palestinian State, which nonetheless itself is perceived as constituting a grave threat to Israel's existence.) More generally, let us assume that a reformulated Zionist account can manage to provide a complete and consistent model. And now, weighing between the two models (which have been made to satisfy the first, or so-called “scientific” condition), let us assume that a functional case is made in favor of the Zionist model over the new account, as being a narrative that could more comfortably accommodate a strategic reformulation of Israel's policy towards Palestinians. The argument, for example, can be made that it was because of the right policy towards Palestinians in the past that Israel has now achieved its security objective and can therefore afford to allow (the defeated) Palestinians their viable state. The militaristic clique can at once feel vindicated and ready to steer Israel in a new direction toward peace. The new account, on the other hand, could be argued to be aggravating, whether to the society as a whole or to its militaristic leaders, and less likely therefore to become a launching pad for a radically new approach towards the Palestinians. In this case, and given the functional criterion, the Isacoff formula would recommend that political scientists favor the Zionist over the new account. And the problem may indeed be resolved more easily—but at a price that political scientists ask historians and their fellow political scientists to pay, namely, not to be too picky about the truth.

In the end, one is compelled to ask whether there is a need, in the face of inconsistency, to opt for a heuristic approach in the first place. Is historical truth, in other words, always and entirely inscrutable—thus requiring replacement or substitution by another standard or measure for validity? Isacoff rightly points out that a political scientist who restricts his selection of historical narratives only to those that present a one-sided account risks arriving at conclusions and outlining theories that will eventually prove to be useless. Clearly, varying and even conflicting narratives have to be studied with equal seriousness. Isacoff's recommendation can be extended even to cover seriously conflicting narratives, such as two opposing “propagandist” accounts, or two opposing worldview accounts, if only in order for the reader to better understand and analyze motivations and perceptions, if not to uncover “hard” data or facts. But one need not for that reason despair of a true account (e.g., of whether Israel pulled the Arabs into a trap in 1967, accompanied by an ingenious public relations campaign). Faced with the evidence that Rabin, for example, was instructed to evict Palestinians from their homes, one need not, as against earlier accounts, decide to become a solipsist. The new evidence simply disproves the earlier account.

There may indeed be cases in which the truth is elusive. A case in point is the Clinton-sponsored Camp David talks four years ago, where several conflicting accounts emerged about what went on and who was to blame. All of those accounts may be true in a sense, but only because each reflects one of several personal perspectives of the event. A historian, one assumes, faced with these various perspectives, must try to discover, like a detective, how the various bits of the puzzle fit together, thus providing posterity, and the political scientist, with an objective bird's eye view of that event.

In choosing his examples from the Israeli-Palestinian case, Isacoff presents us with exceptional material for conducting a foundational debate concerning the function of historians and political scientists. Although restricting his examples to the two aforementioned wars, Israel's historiography, ancient and modern, is replete with examples affecting the ongoing conflict that, if intelligently revisited, can perhaps provide a means for ending that conflict.