13. Aristotle and History

Lucio Bertelli
The title of my paper may wrongly suggest that I am going to discuss Aristotle’s famous—I should perhaps say infamous—comparison in the Poetics between tragedy and history (Poetics 9). In fact, this passage from the Poetics, one of the rare occasions where Aristotle uses the word historia to refer to historiography, is perhaps so famous precisely because of the tendency of modern historians to read it as confirmation of the fact that history was underestimated by Aristotle in comparison with poetry (tragedy as well as epic, if we take Poetics 23 into account) and that the term referred only to the exposition of particular actions or stories of particular individuals (“what Alcibiades did or suffered”) without any overarching significance (to kath’hekaston) (Poetics 9.1451b10–11). It is evident that I cannot escape this subject, but I will save it for the last course of this Aristotelian satura lanx.
A short ‘doxographic’ introduction is necessary to focus my inquiry: the problem of the relationship between Aristotle and history in fact depends more on scholars’ opinions than on Aristotle’s own statements about historia or his modus operandi as a historian – for after the discovery of the Constitution of the Athenians, it became necessary to judge Aristotle not only as a philosopher but also as an historian. And, as is well known, Aristotle’s ability as an historian has often been criticized, from Wilamowitz, who thought that “Aristoteles keine geschichtlicher Forscher ist,” to Jacoby, who questioned the value of Aristotle’s historical work and doubted that he had intended to write a real history. [1] And this, despite the fact that Wilamowitz—and Jacoby implicitly—admitted that the discovery of the Constitution of the Athenians had made it far easier to reconstruct Athens’ constitutional history. This critical judgment at any rate has persevered and gained force in part because of the perception that Aristotle’s historical research about the constitutional development of Athenian democracy is actually an a priori reconstruction based on categories developed in his Politics. [2] Those scholars who consider that Aristotle was an historiographer and that his inquiries do indeed possess some historical value are in fact quite rare: I will mention only those who seem to me most important.
In his still fundamental work, Aristote et l’Histoire. Essai sur la “Politique” (note the subtitle) Raymond Weil approaches the problem by considering quantity—the huge spread and variety of Aristotle’s historical researches—more than quality. How is it possible, he asks, to conclude that Aristotle was not an historian or that he made no contributions to the field of history when the catalogue of his works includes 158 constitutions, Nomima Barbarika, a study of Solon’s axones, accounts of local land claims (Dikaiomata), catalogues of Olympic and Pythian winners, didaskaliai of the tragic and comic competition winners, and historical works on tragedy, on poets, and so on? [3] According to Weil, Aristotle’s ‘condemnation’ of history in Poetics 9 and 23, must be read in the context of his anti-Platonic polemic and not as a final judgment on history itself. However, as Weil argues, Aristotle here makes a distinction between ‘ordinary’, popular historical works (synetheis) and works of great philosophical commitment, such as those of Herodotus and Thucydides: Weil based his conclusion, we should note, on a doubtful reading of the text (Poetics 23.1459a21f.): synetheis ‘popular, customary’ instead of commonly accepted syntheseis ‘structures’. [4]
On several occasions, Kurt von Fritz has also had recourse to reappraise Aristotle’s innovations as an historian. He concludes that, in light of the boundless breadth of his inquiries, Aristotle was the first, indeed the only one in antiquity, to make use of a team of researchers, and he extended his inquiry beyond the traditional bounds of historiography to comprise subjects that had not yet been studied in any systematic way (e.g. biology, philosophical doxography, medicine, and physical sciences), as well as those more closely related to historiography, such as biography. With respect to history in the proper sense, says von Fritz, the unavoidable comparison between history and poetry in the Poetics was only a matter of nuances: poetry was ‘more’ philosophical than history, which is itself concerned with the universal, if ‘less’ so than poetry. [5]
In the most exhaustive work on the subject, Renate Zoepffel rejects von Fritz’s conciliatory solution to Poetics 9, suggesting in fact that the kath’ hekaston ‘particularity’ of historia is not at all compatible with the kath’holou ‘generality’ of poetry and therefore that Aristotle’s concept of historia cannot be compared with ours, nor even with that of Thucydides. [6] Zoepffel’s approach has the merit of setting the famous passage of Poetics 9 in the general framework of Aristotelian research and epistemology in order to explain his idiosyncratic view of historia. But after following Zoepffel’s lengthy theorization of Aristotelian research, her discussion of Aristotle’s conception of history—in her view, Aristotle did not consider history to be cyclical—and her detailed examination of Poetics 9 and 23, we are still unable to answer the question, “Was Aristotle an historian?” We want to know, in other words, whether Aristotle was an historian in accordance with von Fritz’s four requirements of historical work: identification and criticism of traditions and sources, chronological arrangement of facts, explanation of causes, and demonstration of the forces operating within history. [7] Let us see whether we are able to answer this question by relying on the Aristotelian corpus itself.
When Aristotle discusses historia in the Poetics (9.1451b6f. and 23.1459a21f.), he is no doubt thinking about historical works; at 9.1451b2, in fact, he mentions Herodotus as an example of an historikos; his work, he says, would be historical even if it were put into verse. The semantic horizon of the word, historia in Aristotle, however, extends beyond a simple account of genomena ex anthropon ‘the things having happened among men’, beyond even the erga megala ‘great deeds’ of Herodotus and Thucydides; indeed, historia covers all phainomena—including legomena ‘the things said’—of every techne, or science. Closely tied to the Ionic concept of historie, historia in Aristotle in fact plays a fundamental role in his theory of scientific research, which in turn is connected to his theory of the origin and evolution of human knowledge.
At the beginning of the Metaphysics (1.980a28–981b25), Aristotle explains the process of knowledge acquisition through different levels of comprehension: first there is aisthesis ‘perception or sensation’, which is common to all animals; next comes mneme, the memory of sensation, which is peculiar to some animals (man included); belonging to man alone is empeiria, the third stage, which is produced by the fixing of this memory of perception through a process of synthesis; and it is through knowledge of the particular (gnosis ton kath’hekaston) that it is possible to obtain techne and episteme: “art arises when from many notions gained by experience one universal judgment (mia katholou hypolepsis) about a class of objects is produced.” (1.981a6–8; translation by W. D. Ross)
Although it is a lower level of knowledge, Aristotle continues, empeiria is nevertheless significant in the field of praxis (prattein), where the knowledge of the particular is enough to get good results. A theoretician without experience will fail on the practical level; the achievement of scientific knowledge (eidenai) is the result of a combination of knowledge of particular facts (hoti) and knowledge of causes (dioti), and it is accessed only by a person who possesses techne or episteme. Empeiria, then, plays a role—a fundamental role, inasmuch as only empeiria leads to universal judgment (katholou hypolepsis) and allows for causal inference—in the field of the particular (kath’hekaston).
We find this same gnosiological sequence at the end of Posterior Analytics, this time applied to the scientific method (2.19): in the final analysis, the “first principles” (protai archai) of demonstration (apodeixis) come from aisthesis and lead to mneme and empeiria—that is, the fixing of the memory of perception in order to gain unified knowledge, which itself leads to the principle (arche) of techne in the field of becoming and of episteme in the field of being. This inductive process is even more clearly stated in the Prior Analytics:
But most of the principles pertaining to each science are peculiar to it. Consequently, it is the business of experience (empeiria) to give the principles which belong to each subject. I mean for example that astronomical experience supplies the principles of astronomical science: for once the phenomena were adequately apprehended, the demonstrations of astronomy were discovered. Similarly with any other art or science. Consequently, if the attributes of the thing (ta hyparchonta peri hekaston) are apprehended, our business will then be to exhibit readily the demonstrations. For if none of the true attributes of things had been omitted in the historical survey (kata ten historian), we should be able to discover the proof and demonstrate everything which admitted of proof, and to make that clear, whose nature does not admit of proof. (Translation by A. J. Jenkinson, modified)
Aristotle Prior Analytics 1.30.46a19–29
At the end of this chapter, where he discusses the choice of general principles (eklegein tas protaseis) that are necessary for demonstration (apodeixis) or for explicative arguments (phaneron poiein), Aristotle refers to a work on this specific subject, namely his treatise on dialectic better known as the Topics: here, perfectly consistent with the methodological and gnosiological assumptions of the Analytics, the philosopher explains the meaning of eklegein tas protaseis from the point of view of their contents:
As for propositions, they should be selected in a number of ways corresponding to the number of distinctions drawn in regard to the proposition: thus one may first take in hand the opinions held by all or by most men or by the philosophers, i.e. by all, or most, or the most notable of them [. . .] Moreover, all statements that seem to be true in all or in most cases, should be taken as a principle or accepted position [. . .] We should select also from the written handbooks (gegrammenoi logoi) of argument, and should draw up sketch-lists (diagraphai poieisthai) of them upon each several kind of subject, putting them down under separate headings, e.g. ‘On Good’, or ‘On Life’—and that ‘On Good’ should deal with every form of good, beginning with the category of essence [. . .] Of propositions and problems there are—to comprehend the matter in outline—three divisions: for some are ethical propositions, some are on natural philosophy, while some are logical. (Translation by W. A. Pickard-Cambridge)
Aristotle Topics 1.14.105a34–b21
But dialectic is not only a logica inventionis, an art of finding the data or the principles of argumentation; it is also a techne of analysis (exetastike) and proof (peirastike) of the selected data, as Eric Weil correctly emphasized in a famous article. [8] The dialectical process, then, can be applied to all fields of knowledge, in different ways depending on the nature of object at hand. Therefore it can be applied also to the field of objects allos endechomena ‘contingent objects’, and it is not by chance that in the passage from the Topics cited above Aristotle mentions ethical propositions (protaseis ethikai) beside propositions of natural philosophy and of logic (protaseis physikai and logikai). And protaseis ethikai bring us to a field of research overlapping with that of historia qua history, i.e. the field of praxeis. This is demonstrated by the precise reference to the dialectic method that we have seen employed in the Topics—to discover (heurein) and examine the propositions (exetazein tas protaseis)—in the treatises of Aristotle concerned with praxis.
In the methodological prelude on the treatment of “best life and perfect life” in the Eudemian Ethics (1.3.1214b28–1215a8), Aristotle stresses the necessity of “investigating the opinions” (episkopein tas doxas) on the subject, but he “scrutinizes” (exetazein, a term that points to the relationship of this process to the dialectic method) only the opinions of wise men, not those of “boys, sick, and mad people.” In the Rhetoric (2.22.1396a5f.), in a passage that treats “discussions or reasonings about politics or every subject”—with a clear reference to the Topics about the necessity of having at hand “a choice of probable and more suitable premises” for each subject—Aristotle says that “it is necessary to possess the elements of the argument (hyparchonta), all or part of them.” It is true that in the following examples, divided as they are according to the circumstances of discourse (deliberative, encomiastic, or invective), the philosopher does not separate historical hyparchonta (Athens and Sparta) from mythical ones (gods, Achilles); but when we must deliberate about whether or not to go to war, the ‘premises’ from which our conclusions come are based strictly on knowledge that is factual (the availability of military forces, economic resources, alliances, and hostilities) and ‘historical’ (past wars and their consequences). When dealing with discourses of praise and condemnation, moreover, Aristotle recalls not only the mythical precedent of the Heraclids but also the historical battles of Salamis and Marathon and, in the case of invective (psogos), the enslavement of former allies during the Persian Wars (such as the Aeginetans and Potidaians), all of which requires a knowledge of history.
The use of examples from the past (paradeigmata) is a technique explained, as Roberto Nicolai has shown in his excellent book, by Aristotle’s principle that “future cases are usually similar to past ones” (Rhetoric 2.20.1394a5). [9] If we replace Aristotle’s cautious hos epi poly ‘for the most part, usually’ here with the more absolute kata to anthropinon, ‘according to human nature’, we in fact obtain Thucydides’ justification of historiography (1.22.4). In a frequently quoted chapter of the Rhetoric, moreover, where Aristotle lists the most popular subjects of political discourse—poroi ‘ways and means’, peace and war, phylake ‘defence’ of the territory, import and export, and legislation—he demands not only that the political speaker have personal experience in the field but also that he be historikos, ‘a searcher’ of other men’s solutions and facts of the past. This latter kind of knowledge is particularly necessary in legislation (nomothesia):
But while he must, for security’s sake, be able to take all this into account, he must before all things understand the subject of legislation; for it is on a country’s laws that its whole welfare depends. He must, therefore, know how many different forms of constitution there are; under what conditions each of these will prosper and by what internal developments or external attacks each of them tends to be destroyed. When I speak of destruction through internal developments I refer to the fact that all constitutions, except the best one of all, are destroyed both by not being pushed far enough and by being pushed too far. Thus, democracy loses its vigour, and finally passes into oligarchy, not only when it is not pushed far enough, but also when it is pushed a great deal too far [here Aristotle has recourse to one of his favorite analogies, a nose that is either too aquiline or too snub]. It is useful, in framing laws, not only to study the past history of one’s own country (ek ton paralelythoton theoria, i.e. historical knowledge), in order to understand which constitution is desirable for it now, but also to have a knowledge (eidenai) of the constitutions of other nations, and so to learn for what kinds of nation the various kinds of constitution are suited. From this we can see that books of travel (ges periodoi) are useful aids to legislation, since from these we may learn the laws and customs of different races. The political speaker will also find the researches of historians useful. [10] But all this is the business of political science and not of rhetoric. (Translation by W. Rhys Roberts, modified)
Aristotle Rhetoric 1.4.1360a18-38
Aristotle’s conclusion here is perfectly correct, inasmuch as what he lists as necessary for nomothesia well summarizes the contents of the Politics and corresponds for the most part to his proposed plan for that treatise appended to the end of Nicomachean Ethics (10.9). I will avoid the insoluble—or almost insoluble—problem of the chronology of Aristotle’s treatises, aside from pointing out that the collected constitutions (synegmenai politeiai, describing 158 politeiai), which clearly underlie the concluding passage of the Nicomachean Ethics about the causes of destruction and salvation of cities and their politeiai, did not exist when Aristotle wrote the first book of the Rhetoric (usually considered to be one of his earliest works). The same can be said of his treatises on the laws of non-Greek peoples. It is, of course, possible to say that in the chapter of the Rhetoric cited above, Aristotle is speaking about a (political) speaker’s socio-political culture, not explaining his own research project. Nevertheless, if the works on which he based the argumentation of the Politics, not to mention the Politics itself, had already been written, we would expect Aristotle to have alluded to them in the Rhetoric, by way of a source citation, for example.
In fact, in the Rhetoric Aristotle mentions only works of others: the descriptions of the earth (ges periodoi) of the ancient Ionic tradition (he cites Scylax of Caryanda in the Politics) and anonymous historical works (praxeis). And his reference to politics as the techne of legislation makes sense also according to a broader point of view concerning the very structure of the Politics as a treatise. As Raymond Weil astutely observed in his article on Aristotle’s Politics, a palinode to his conclusions in Aristote et l’histoire, if it is true that Aristotle’s political philosophy depends on his historical knowledge—and the conclusion of the Nicomachean Ethics seems to me to be the final proof of this—the reciprocal is true as well, i.e. that “the historiographer Aristotle remembers himself to be something other than an historian.” [11] If we want to understand Aristotle as historian, then, we must make a synthesis of the two positions.
Up to this point in my argument, the assertions, methodological principles, and practical applications displayed by Aristotle as a collector of phainomena (including past events), that is as an historikos in the larger sense of the word, can be used more as evidence for his views on the “utility of history”—intended as a “mine of informations to exploit” as said by Nicolai [12] —than to answer the question that we originally posed, namely, “Was Aristotle able to construct an historical argument or an historical récit in accordance with von Fritz’s criteria?” It would, of course, be possible to answer this question by analyzing the only work of Aristotle’s (problems of authorship apart) that is ostensibly historical, namely the Constitution of the Athenians. But, it is precisely the author of the Constitution of the Athenians who has been accused of not being a proper historian. To defend Aristotle as an historian, then, it will be necessary to look elsewhere.
Unfortunately, we cannot rely on the evidence from the fragments of the other 157 constitutions—in fact, only 148 constitutions, Athens included, are attested—because they tend to be antiquarian remarks made by scholars interested in curious particulars and not in the historical quality of the information they cite. No help comes from Herakleides Lembus’ Excerpta. If Ptolemy VI had to learn how the Athenian constitution worked, he would have had better success consulting Aristotle’s still extant originals than his secretary’s summaries: certainly it was not enough for him to be informed that in Athens there were astynomoi overseeing the measures of balconies on the streets, that the Eleven took care of prisons, that there were nine archons, among which there was one—the Basileus—who dealt with matters of religion and war, or that the six thesmothetai swore not to take bribes, or else they would “set up a golden statue.” Not to mention other amenities of the summary. And, in my opinion, the summary of the Spartan constitution is no better. . .. [13] I would prefer to examine as examples of historical reasoning the ‘unintentional’ evidence contained in Aristotelian works whose aim is not to provide an account of historical events.
I will begin with Aristotle’s ‘cameo’ of cultural history in the first book of the Metaphysics, a passage that immediately follows his theory of the epistemological process (aisthesis-mneme-empeiria-techne or episteme) and that aims to confirm this sequence with historical facts:
At first he who invented any art whatever that went beyond the common perceptions (aistheseis) of man was naturally (eikos!) admired by men, not only because there was something useful (chresimon) in the inventions, but because he was thought wise and superior to the rest. But as more arts were invented, and some were directed to the necessities of life (anankaia), others to recreation (diagoge), the inventors of the latter were naturally always regarded as wiser than the inventors of the former, because their branches of knowledge (epistemas) did not aim at utility. Hence when all such inventions were already established, the sciences which do not aim at giving pleasure or at the necessities of life were discovered, and first in the places where men first began to have leisure (schole). This is why the mathematical arts [14] were founded in Egypt; for there the priestly caste was allowed to be at leisure. (Translation by W. D. Ross)
Aristotle Metaphysics 1.981b14–26
In this encapsulation of cultural history, Aristotle scrupulously follows the rules of scientific inquiry that he set out in the Analytics, namely the relationship between hoti (here, the discovery of mathematics in Egypt) and dioti (the theory of the evolution of knowledge), but the reconstruction, despite its brevity, also satisfies von Fritz’s conditions of historical writing: (1) Collection and criticism of sources. The source for the discovery of mathematics in Egypt is Herodotus 2.109, supplemented—as seems clear from scholia—with a passage from Isocrates (Busiris 21). We cannot exclude the possibility that there were other influences, as well, a passage from Democritus’s Mikros Diakosmos (68 B 5 D.K.), for example, regarding inventions for recreation (pros diagogen) and for pleasure (pros hedonen). But we can note that Plato’s mythical version of the Egyptians’ contributions (Phaedrus 274) goes completely unnoticed. (2) Chronological ordering. The sequence of knowledge is chronologically arranged, emphasized by the adverbial progression proton-eita; in fact, the reference to the Egyptians is itself a chronological marker, already evident in Herodotus. (3) Causal connections. This is quite clear in the teleology for movers of inventions: first there is use/necessity (chreia/ananke), then recreation (diagoge) and leisure (schole), which is the prerequisite of techne/episteme. This last causal connection clearly demonstrates (4) the “evolutionary forces” in action.
The Egyptians evidently also play a part in the second piece of evidence that supports my contention about Aristotle’s validity as an historian, namely a short historical and cultural excursus from the Metaphysics about the divinity of heavenly bodies and the origin of this belief:
Our forefathers in the most remote ages (para ton archaion kai pampalaion) have handed down (paradedotai) to their posterity a tradition, in the form of a myth (en mythou schemati kataleleimmena), that these bodies are gods, and that the divine encloses the whole of nature. The rest of the tradition has been added later in mythical form with a view to the persuasion of the multitude and to its legal and utilitarian expediency (pros ten peitho ton pollon kai pros ten eis tous nomous kai to sympheron chresin); they say these gods are in the form of men or like some of the other animals, and they say other things consequent on and similar to these which we have mentioned. But if one were to separate the first point from these additions and take it alone—that they thought the first substances to be gods—one must regard this as an inspired utterance, and reflect that, while probably (kata to eikos) each art and each science has often been developed as far as possible and has again perished, these opinions, with others, have been preserved until the present like relics (leipsana) of the ancient treasure. Only thus far, then, is the opinion of our ancestors (patrios doxa) and of our earliest predecessors (proteron) clear to us. (Translation by W. D. Ross)
Aristotle Metaphysics 12.8.1074a38–b16
En mythou schemati kataleleimmena ‘a tradition in the form of a myth’, leipsana ‘relics, remains’: we feel Thucydides, perhaps, behind these words. In fact, if we apply Aristotle’s argument to the material conditions of Greece at its origin, taking also into account his revaluation of mythologountes ‘mythologizers’ or leipsana as evidence for the past, we would approximate some passages from Thucydides’ Archaiologia.
In this short discussion of the origin of the doxa ‘belief’ about the divinity of stars, Aristotle not only reveals his global view of history—the repetition in time of human experiences, which is a Thucydidean principle—but also criticizes the paradedomena ‘the views handed down’ of much of the ancient tradition, resorting without problems to rather dangerous prompters as Critias, to whom, I believe, Aristotle points with the phrase, pros ten peitho . . . sympheron chresin ‘with a view to the persuasion. . .and utilitarian expediency’.
There is another text that explicitly demonstrates Aristotle’s ability to place the historical kath’hekaston into a causal chain—and what a causal chain!—where it serves as evidence for the necessary and natural evolution of events. This ‘unintentional testimony’ comes from a treatise completely extraneous to the history of human events, the Meteorologica. In a passage from Book 1, Aristotle explains why lands and seas change over time, why at one time there may be sea where once there was land, and vice versa, and why rivers sometimes appear and sometimes dry up (1.14). All of these things, he says, depend on the alternate cycle of hot and cold and their inherent qualities, dry and humid. All of these things happen according to a cyclical order, but this order concerns not the whole kosmos, but only those parts of earth that change cyclically in a sort of seasonable cycle, similar to the biological one of growth and senility. This cycle manifests itself differently in different parts of earth because physical bodies—that is, terrestrial elements—change differently from place to place, in accordance with the course of the sun (periphora), which sometimes moves away and sometimes moves close to earth. Animal bodies, on the contrary, change simultaneously in all their elements. Therefore there are not only seasons of the year but also ‘seasons’ of the earth, with an alternation between hot and cold and their consequences.
Up to this point, Aristotle’s physical theory is totally consistent with his astronomical conceptions and offers a sufficient explanation of the phenomenon. Next, Aristotle introduces a comparison between human and geological time in order to show why men cannot directly perceive geological changes. It is possible, he insists, to perceive this change by taking into account the evidence offered by human history, provided we examine it on the basis of dioti, the necessary sequence of phenomena.
But the whole vital process of the earth takes place so gradually and in periods of time which are so immense compared with the length of our life, that these changes are not observed, and before their course can be recorded from beginning to end whole nations perish and are destroyed. Of such destructions the most utter and sudden are due to wars; but pestilence or famine cause them too. Famines, again, are either sudden and severe or else gradual. In the latter case the disappearance of a nation is not noticed because some leave the country while others remain; and this goes on until the land is unable to maintain any inhabitants at all. So a long period of time is likely to elapse from the first departure to the last, and no one remembers and the lapse of time destroys all record even before the last inhabitants have disappeared. In the same way a nation must be supposed to lose account of the time when it first settled in a land that was changing from a marshy and watery state and becoming dry. Here, too, the change is gradual and lasts a long time and men do not remember who came first, or when, or what the land was like when they came. This has been the case with Egypt. Here it is obvious that the land is continually getting drier and that the whole country is a deposit of the river Nile. But because the neighbouring peoples settled in the land gradually as the marshes dried, the lapse of time has hidden the beginning of the process. However, all the mouths of the Nile, with the single exception of that at Canopus, are obviously artificial and not natural. And Egypt was nothing more than what is called Thebes, as Homer, too, shows, modern though he is in relation to such changes. For Thebes is the place that he mentions; which implies that Memphis did not yet exist, or at any rate was not as important as it is now. That this should be so is natural, since the lower land came to be inhabited later than that which lay higher. For the parts that lie nearer to the place where the river is depositing the silt are necessarily marshy for a longer time since the water always lies most in the newly formed land. But in time this land changes its character, and in its turn enjoys a period of prosperity. For these places dry up and come to be in good condition while the places that were formerly well-tempered some day grow excessively dry and deteriorate. This happened to the land of Argos and Mycenae in Greece. In the time of the Trojan wars the Argive land was marshy and could only support a small population, whereas the land of Mycenae was in good condition (and for this reason Mycenae was the superior). But now the opposite is the case, for the reason we have mentioned: the land of Mycenae has become completely dry and barren, while the Argive land that was formerly barren owing to the water has now become fruitful. Now the same process that has taken place in this small district must be supposed to be going on over whole countries and on a large scale. (Translation by E. W. Webster)
Aristotle Meteorologica 1.14.351b8–352a17
R. Zoepffel also discusses this extraordinary passage of the Meteorologica, but she uses it to question whether or not Aristotle had a cyclic view of the world and history, and does not consider it as evidence, which it surely is, of Aristotle’s noteworthy ability to construct an historical account. [15] He starts from the premise that there is a disproportion of time between geological and human events, an observation based on the factual data of the shorter duration of people’s life and the mechanism of nations’ decay (phthorai)—due in part to chance (polemoi ‘wars’, nosoi ‘sicknesses’), in part to necessity (aphoriai ‘famines’, caused by the geological cycle of hot and cold)—and proceeds to the analysis of symbebekota—Egypt, Argos and Mycenae. Through the criticism (exetasis) of signs (tekmeria)—the mouths of Nile, the name of Egypt in Homer, the comparison between ancient and contemporary conditions of the two Greek cities—he confirms (peira) the premise. We recall, perhaps, that Thucydides also alludes to Argos and Mycenae in his Archaiologia.
As further evidence that Aristotle fulfils the criterion of the “collection and criticism of sources,” note his ‘proof’ at the end of the same chapter (Meteorologica 1.14.352b24–31) that serves to validate his theory of local geological changes (metabolai), namely the story of the canal dug from Bubastis to the coast of the Red Sea (the Erythrean Sea). Herodotus also records this event (2.158), but his version does not seem to lie behind that of Aristotle, for Aristotle ascribes the canal not to Nekos, as had Herodotus, but to Sesostris. Furthermore, Aristotle claims that the project was abandoned by Sesostris (and again by Darius) out of fear that salty waters would seep into the Nile and spoil the river, since the sea was higher than the land. Herodotus, on the other hand, had supposed that Nekos ceased digging because an oracle predicted that the canal would end up benefiting the Barbarians and the work would be completed by Darius. We cannot tell what source Aristotle was using here, but it was certainly not his favourite one, Herodotus.
If we want to take the Politics into consideration, a work that befits the formula “sans histoire, pas de matière pour la politique,” a paraphrase of what Eric Weil once said about the Topics and dialectical syllogism, [16] we find that historical events (genomena) are cited not only to show indications (martyria) or signs (semeia) of constitutional changes but also as evidence for the evolution of political and sociological institutions according to a logic not based on chance but subject to strict causal connections (see, for example, 1.8 on nourishment (trophe) and nations; 3.13 on ostracism; 3.14 on forms of monarchy; and 7.11 on forms of cities). Think, too, of the sequence, monarchy-archaic polity, at Politics 4.13, which is caused by the evolution of military forces (cavalry to hoplites) and by the widening of political participation to include hoplites.
Having explored Aristotle’s use of the concept of historia, let us conclude by returning to the starting point of our inquiry, to the famous chapters of the Poetics and to Aristotle’s allegedly ‘scandalous’ underestimation of history. As should be evident from our survey of the role of historia as a preliminary inquiry into the kath’hekasta in Aristotle’s research method, valid—as Aristotle makes clear—for every techne and episteme, we can conclude that in Poetics 9 there is no limit put on the function of historia and that the philosopher in no way claims that the historical kath’hekasta cannot be used to construct arguments that are philosophoterai (causal according to verisimilitude and necessity). In fact, in the examples we have examined he does precisely this. It is certainly true that the example chosen to illustrate the historical ‘particular’ in Poetics 9 is a little trifling—what Alcibiades did and what he suffered—but we must consider that Aristotle is here dealing with the actions (praxeis) and speeches (logoi) of tragic characters and so the historical object must be analogous. The ‘scandal’, if it is a real scandal, is the way in which Aristotle describes historiai:
Epic compositions (syntheseis) will differ in structure from historical compositions (historiais), which of necessity present not a single action, but a single period, and all that happened within that period to one person or to many, little (hos etuchen) connected together as the events may be. For as the sea-fight at Salamis and the battle with the Carthaginians in Sicily took place at the same time, but did not tend to any one result, so in the sequence of events, one thing sometimes follows another, and yet no single result is thereby produced. (Translation by S. H. Butcher)
Aristotle Poetics 23.1459a21-29
Several interpretations are possible: perhaps Aristotle has not yet written the Constitution of the Athenians; perhaps, alternatively, we should eliminate from the Constitution of the Athenians the passage in which causal connections explain the chronological sequence of eleven changes (metabolai) in the Athenian constitution (Constitution of the Athenians 41). But there is a third option, too: the philosopher may be thinking here of another kind of history, a pure chronicle, similar to the precursors to the The Parian Chronicle (Marmor Parium).
In my opinion, it is not necessary to change syntheseis (structures, compositions) into synetheis (sc. historias: ordinary, popular histories) or to embark on syntactical tightrope walks—as does Weil—to justify this lesson in the text. Aristotle is here comparing the supreme Homer to the epical poetasters who make mythical chronicles and not unitary epical compositions whose narrative has a beginning, a middle and an end; in the same way, he elliptically compares the unitary epical action to the historical chronicle. In so doing, he in fact suggests that there is another kind of history, quite different from the simple chronicle. I am certainly aware that I am using an argumentum ex silentio, which may perhaps convince, but which lacks scientific rigor. But, if we do not accept this explanation, we must assume either that Aristotle is using historical facts here in a way that is inconsistent with his usual method, or that, in the Poetics, he has reduced history to ‘random’ chronicle.
I am inclined without hesitation to give a positive answer to the question with which I began: “Was Aristotle able to write history?” Aristotle’s history is a peculiar history based on a theory of scientific research and knowledge that we do not usually find in historians, but we cannot deny that Aristotle’s circumstantial processes are very similar to those of Thucydides. In conclusion, it is hard to deny that a person who affirms that we must rely more on observation than on theory, when the facts have not yet been satisfactorily verified, and that we must rely on theory only when it matches facts already ascertained (see De Generatione Animalium 760b30–33) has the forma mentis of an historian, of an Erforscher. Aristotle certainly did not lack any of the principles necessary for discovering the causes of events.


Bertelli, L. 2004. “La Sparta di Aristotele: un ambiguo paradigma o la crisi di un modello?” In Sparta fra tradizione e storia, ed. R. Vattuone, 9–71. Bologna.
Day, J., and M. Chambers, eds. 1962. Aristotle’s History of Athenian Democracy. Berkeley.
Gallavotti, C. 1974. Aristotele. Dell’arte poetica. Verona.
Gudeman, A. 1934. Aristoteles. Perì Poiêtikês. Berlin.
Jacoby, F. 1949. Atthis. The Local Chronicles of Ancient Athens. Oxford.
Fritz, K. von. 1936. “Herodotus and the Growth of Greek Historiography.” Transactions of the American Philological Association 67:315–340.
———. 1958a. “Die Bedeitung des Aristoteles für die Geschichtsschreibung.” In Histoire et Historiens dans l’antiquité, ed. K. Latte, 85–128. Entretiens Hardt IV. Vandoeuvres-Genève.
———. 1958b. Aristotle’s Contribution to the Practice and Theory of Historiography. Berkeley.
Nicolai, R. 1992. La storiografia nell’educazione antica. Pisa.
Weil, E. 1951. “La place de la logique dans la pensée aristotélicienne.” In Revue de métaphysique et de morale 56:283–315.
Weil, R. 1960. Aristote et l’histoire. Essai sur la ‘Politique’. Paris
———. 1965. “Philosophie et histoire. La visione de l’histoire chez Aristote.” In La ‘Politique’ d’Aristote, ed. R. Stark et al., 159–189. Entretiens Hardt XI. Vandoeuvres-Genève.
Wilamowitz-Moellendorff, U. von. 1966. Aristoteles und Athen I-II. 2nd ed. Berlin. Orig. pub. Berlin 1893.
Zoepffel, R. 1975. Historia und Geschichte bei Aristoteles. Abhandlungen der Heidelberger Akademie der Wissenschaften, Philosophisch-historische Klasse 2. Heidelberg.


[ back ] 1. Wilamowitz 1966:i373; Jacoby 1949:210.
[ back ] 2. See e.g. Day and Chambers 1962.
[ back ] 3. Weil 1960.
[ back ] 4. Weil 1960:163–178; for the text, see Gudeman 1934:388f. and Gallavotti 1974:88.
[ back ] 5. Fritz 1958a; 1958b.
[ back ] 6. Zoepffel 1975.
[ back ] 7. Fritz 1936:315.
[ back ] 8. Weil 1951.
[ back ] 9. Nicolai 1992:42–46.
[ back ] 10. Rhys Roberts’s translation in this point is not correct. It must be: “and an aid to political proposals (politikai symboulai) comes from the researches on the authors of historical works (historiai ton peri tas praxeis graphonton).”
[ back ] 11. Weil 1965:161.
[ back ] 12. Nicolai 1992:47.
[ back ] 13. Herakleides’ summary of Aristotle’s Constitution of the Lacedaimonians is very much along the same lines; this is not, however, to say that there were no other benefits to Herakleides’ work. See Bertelli 2004.
[ back ] 14. Aristotle uses technai and not epistemai; probably the reason of this choice is hidden in the source or sources he is using here.
[ back ] 15. Zoepffel 1975:45–51.
[ back ] 16. Weil 1951:292.