Errors in Arrian
by A. B. Bosworth
The Classical Quarterly, New Series, Vol. 26, No. 1 (1976), pp. 117-139
Published by Cambridge University Press on behalf of The Classical Association
1976
NOTICE: THIS WORK MAY BE PROTECTED BY COPYRIGHT
YOU ARE REQUIRED TO READ THE COPYRIGHT NOTICE AT THIS LINK BEFORE YOU READ THE FOLLOWING WORK, THAT IS AVAILABLE SOLELY FOR PRIVATE STUDY, SCHOLARSHIP OR RESEARCH PURSUANT TO 17 U.S.C. SECTION 107 AND 108. IN THE EVENT THAT THE LIBRARY DETERMINES THAT UNLAWFUL COPYING OF THIS WORK HAS OCCURRED, THE LIBRARY HAS THE RIGHT TO BLOCK THE I.P. ADDRESS AT WHICH THE UNLAWFUL COPYING APPEARED TO HAVE OCCURRED. THANK YOU FOR RESPECTING THE RIGHTS OF COPYRIGHT OWNERS.
Arrian is regarded as the most authoritative of the extant sources for the reign of Alexander the Great. It is his work that is usually chosen to provide the narrative core of modern histories, and very often a mere reference to 'the reliable Arrian' is considered sufficient to guarantee the veracity of the information derived from him. What gives Arrian his prestige is his reliance on contemporary sources, Ptolemy and Aristobulus. It is recognized that Arrian's narrative is based primarily upon Ptolemy, and, as long as Ptolemy is regarded as an impeccable mine of facts for Alexander's reign and Arrian's work is accepted as a faithful reproduction of Ptolemy, the Anabasis Alexandri stands out as a uniquely authoritative record of Alexander's reign. Indeed Arrian's text has been thought to retain the original wording of Ptolemy, itself supposedly copied from a hypothetical court journal kept by Eumenes of Cardia. In that case Arrian's Anabasis is only two removes from the actual archives of Alexander.1 The first of these tenets of belief, the factual reliability of Ptolemy, has recently been under attack. Not only has it been virtually disproved that Ptolemy constructed his history from archival material, but it appears that he inserted his own propaganda to exaggerate his personal achievements under Alexander and to discredit those of his rivals.2 It is the second tenet, the faithfulness of Arrian's reproduction of his sources, that I wish to examine, and I shall try to prove that Arrian was prone to the errors of misunderstanding and faulty source conflation that one would expect in a secondary historian of antiquity.
We should first examine what Arrian says about his historiographical aims, both in the Anabasis and elsewhere. In the Anabasis Arrian makes it quite plain that his work is designed as a literary showpiece. Alexander's achievements, he says, have never been adequately commemorated in prose or verse. The field is therefore open for him to do for the Macedonian king what Pindar had done for the Deinomenid tyrants and Xenophon for the march of the Ten Thousand.3 Throughout his work Arrian is proudly conscious of his stylistic mastery, and he assumes a wide public. He openly claims the first place in Greek literature and at the outset he states that his history will stand comparison with anything previously written about Alexander.4 Arrian is of course concerned with veracity, and he seems convinced that his choice of sources guarantees the truth of his account. Indeed a by-product of his work will be to stop the transmission of popular fantasies from generation to generation.5 Literary considerations, however, are paramount. Arrian claims superiority to all Alexander historians, not merely the inferior stylists of the early Hellenistic period but also writers of his own age. Arrian's fellow countryman, Dio Cocceianus of Prusa, had written an eight-volume work [x], and we can be sure that it had an encomiastic [complimentary] bias.6 Alexander's achievements had, then, been praised in prose by the leading Bithynian orator and stylist of the previous generation. Arrian intended to do better.
It is unwise to exaggerate the importance in Arrian's eyes of his work on Alexander. The subject attracted him, he says, because it was a field worthy of his talents.7 He does not imply that his preoccupation with Alexander was of long standing. When he says that for him his country, family, and public offices have always been [x], he is referring not so much to his specific work on Alexander (i.e. 'this treatise') but to the totality of his literary production (i.e. 'these writings of mine'). The parallels from Arrian's usage elsewhere suggest that when he refers to a single monograph he describes it in the singular as a [x], not in the plural.8 Arrian speaks as a writer assured of his literary mastery and reputation; he is actually doing a service for Alexander.9 Now we know indirectly that Arrian did not regard his work on Alexander as his major literary task. In the preface to the Bithyniaca he claimed that it was the history of Bithynia that occupied his attention from the outset of his historical career. The preparation had taken an excessive time because of his inexperience in the field, and in the interim he had published monographs on Dion and Timoleon and also his work on Alexander.10 The history of Alexander was in a sense preparatory, but hardly a stylistic exercise, as Schwartz thought. In the Anabasis Arrian proudly asserts that he heads the field of Greek literature, and nothing in his writing suggests stylistic immaturity. Arrian stresses in the Bithyniaca that it was the preparation ([x]) that was time-consuming. What this implied is clear from Lucian's treatise on historiography: the primary work of collecting information and shaping it in the correct proportions in a rough first draft, or [x]. The stylistic embellishment was a later, more important, task.11 The history of Bithynia involved a vast coverage from mythological times until the death of Nicomedes IV Philopator in 74 B.C. There was a mass of material, mostly compiled during the Hellenistic period,12 and the process of selection and assimilation must have been difficult. It is hardly surprising that Arrian turned to more limited periods of non-contemporary history in order to gain expertise in the creating of a unitary, seamless narrative from disparate primary sources.
We can now understand the peculiarity of the Anabasis. Stylistically it is the work of a mature and skilful writer, steeped in the models of the classical period and with all the figures of the schools at his command. At the same time Arrian is relatively inexperienced in the selection and juxtaposition of material from his primary sources, and it is here that we may expect to find misunderstandings due to over-hasty reading and doublets arising from imperfect conflation of the two narratives. Such errors are commonplace in Livy, an author who, like Arrian, wrote non-contemporary history,13 and it would be naive to suppose that Arrian was free from error at this early stage of his historical development. In what follows I shall illustrate the two most characteristic types of error, misunderstanding of a single source and imperfect reconciliation of variant traditions. I shall then approach two important problems of Alexander's reign, one new and one old, and show how the whole aspect of the question can change once it is accepted that Arrian's technique is fallible.
I
In the autumn of 335 Alexander returned to Macedonia after the destruction of Thebes and the demand for the surrender of the Athenian statesmen. Once in Macedon, says Arrian, he conducted the sacrifice to Olympian Zeus which King Archelaus had instituted and held the Macedonian Olympia at Aegae ([x]).14 Arrian explicitly locates the Olympia at the old capital of Aegae, and he is clearly wrong. Every other source which refers to the Macedonian Olympia stresses that they were held at Dium, a city well to the south of Aegae, less than a mile from the foot of Mt. Olympus.15 The Olympia were held regularly at Dium, for, when Scopas and his Aetolians devastated the city in 219 B.C., among the items destroyed were the installations for the festival.16 We cannot even reckon on the possibility of an irregular celebration at Aegae in 335, for Diodorus also describes the return to Macedonia and says explicitly that at Dium Alexander carried out elaborate sacrifices and held the nine-day festival in honour of Zeus and the Muses which Archelaus had established.17 This festival for Zeus and the Muses was in fact the Macedonian Olympia; that is plainly stated by the scholia to Demosthenes. There can be no doubt that in 335, as in other years, the Olympia were held at Dium. Dio Cocceianus mentions a similar tradition that after the battle of Chaeronea Philip and Alexander sacrificed to the Muses at Dium and held the Olympian festival.18 The tradition is consistent and points to an error in Arrian.
The error did not arise from source conflation, for immediately after his reference to the Olympia held at Aegae Arrian mentions a variant tradition that Alexander held a festival in honour of the Muses.19 The festival was of course the Olympia at Dium. Arrian found references in his two sources to the Olympia and to the festival for the Muses, and he inferred erroneously that they were different ceremonies. The reference to the Olympia at Aegae must be taken from a single source, and the error is either one of misunderstanding by Arrian or a mistake on the part of his primary source. If that source was, as is generally thought, Ptolemy, the error is incomprehensible. Even if the passage derives from Aristobulus, serious difficulties remain, for Aristobulus was a resident of Cassandreia in Macedonia, and he ought to have been well informed about the Olympia. The misunderstanding is most probably Arrian's own. Now it is striking that Archelaus' name appears in Arrian only in the context of the sacrifice to Olympian Zeus; in the Diodoran tradition he is associated rather with the establishment of the Olympia. It seems implausible that a sacrifice to Olympian Zeus was first instituted by Archelaus, who reigned as late as 413-399 B.C. As the chief god of the Macedonian pantheon and the ultimate ancestor of the Argead house one would expect Zeus to have been honoured with sacrifices from time immemorial.20 If, however, we accept the alternative tradition that it was the Olympia that Archelaus established, we can explain the possible origin of the error. The Olympia were celebrated some time after the Great Mysteries at Athens,21 that is, after the latter part of September. Very probably they inaugurated the Macedonian New Year, which fell roughly in the middle of October; and the first month of the Macedonian year was Dios, the month of Zeus.22 It might be the case that until the reign of Archelaus the New Year had begun with a ceremonial sacrifice to Zeus in the old capital, Aegae. Archelaus then transferred the sacrifice to Dium and associated it with the newly established festival for the Muses, the Olympia. Arrian's source perhaps gave a brief history of the New Year celebrations, mentioning the original sacrifice at Aegae and the establishment of the Olympia by Archelaus. If Arrian had been working hurriedly, he might easily have misunderstood the passage and supposed that Archelaus established the sacrifice and that the Olympia was held at Aegae. The passage is historically unimportant, but it is a clear example of an error of misunderstanding and raises the possibility that there may be similar mistakes in the account of more serious matters.
A more important issue is raised by Arrian's account of Alexander's administrative measures at Susa in December 331 (3. 16. 9). The satrap of Susiane who had held office under Darius was confirmed in his position under the supervision of a Macedonian general, Archelaus son of Theodorus. The citadel of Susa was placed under a separate commander, named by Arrian [x]: 'Mazarus, one of the hetairoi'.
The Companions (Greek: ἑταῖροι [heˈtairoi̯], hetairoi) were the elite cavalry of the Macedonian army from the time of king Philip II of Macedon, achieved their greatest prestige under Alexander the Great, and have been regarded as the first or among the first shock cavalry used in Europe. Chosen Companions, or Hetairoi, formed the elite guard of the king (Somatophylakes).
-- Companion cavalry [Hetairoi], by Wikipedia
Now Mazarus is a suspiciously Iranian-sounding name, reminiscent of Mazares the Mede mentioned by Herodotus,23 and the aftermath of Gaugamela is a remarkably early time to find an oriental assimilated into the Macedonian court elite.
The Battle of Gaugamela (/ˌɡɔːɡəˈmiːlə/; Greek: Γαυγάμηλα), also called the Battle of Arbela (Greek: Ἄρβηλα), was a battle that took place in 331 BC between the forces of the Army of Macedon under Alexander the Great and the Persian Army under King Darius III. It was the second and final battle between the two kings, and is considered to be the final blow to the Achaemenid Empire, resulting in Alexander's complete conquest of the empire.
The fighting took place in Gaugamela, which literally meant "The Camel's House", a village on the banks of the river Bumodus. The area today would be considered modern-day northern Iraq. Alexander's army was heavily outnumbered & modern historians say that "the odds were enough to give the most experienced veteran pause". Despite the overwhelming odds, Alexander's army emerged victorious due to the employment of superior tactics and the clever usage of light infantry forces. It was a decisive victory for the League of Corinth, and it led to the fall of Achaemenid Empire [First Persian Empire] and of Darius III.
The Battle of Gaugamela is regarded as the climax of a prolonged war between East and West; a war that had been intermittently waged for more than a century and a half. The victory also resulted in, and solidified, Alexander's position as the standing king of Asia.
-- Battle of Gaugamela, by Wikipedia
It is true that Persians who had surrendered in the early years of the campaign, men like Mithrines and Amminapes, were honoured with satrapies in the year following Gaugamela,24 but there is no suggestion that they had been made hetairoi.
Satraps (/ˈsætrəp/) were the governors of the provinces of the ancient Median [Iranian] and Achaemenid [First Persian] Empires and in several of their successors, such as in the Sasanian [Neo-Persian] Empire and the Hellenistic empires. The satrap served as viceroy to the king, though with considerable autonomy. The word came to suggest tyranny or ostentatious splendour.
-- Satrap, by Wikipedia
Berve was therefore forced to conclude that, despite his name, Mazarus was a prominent Macdonian.25 Curtius, however, gives a version of the Susa appointments which is fuller than Arrian's. Abulites' confirmation as satrap is mentioned, as is the appointment of Archelaus with an army 3,000 strong.26 Curtius adds that a certain Callicrates was put in charge of the treasuries, a piece of circumstantial information which has not been challenged. He also mentions the citadel commander and his garrison of 1,000 veterans, but he names him not Mazarus but Xenophilus. Now Xenophilus is a known figure. In 317/16 he was treasurer and garrison commander of the citadel of Susa, and in that role he supported Eumenes against Antigonus.27 According to Curtius, Xenophilus was appointed by Alexander as early as 331. Faced with the undeniable fact that Xenophilus was indeed citadel commander of Susa later on, scholars have usually assumed that Curtius' source erroneously anticipated his appointment; he was, it is argued, Mazarus' successor, appointed late in Alexander's reign.28
It is more likely that Arrian is wrong. Curtius' record of the appointments at Susa, we have seen, is far fuller than that of Arrian. In general Arrian's narrative of the period between Gaugamela and the firing of Persepolis is extremely compressed and scrappy. The so-called vulgate tradition gives much fuller detail, particularly where appointments are concerned.
The History of Alexander, also known as Perì Aléxandron historíai, is a lost work by the late-fourth century BC Hellenistic historian Cleitarchus, covering the life and death of Alexander the Great. It survives today in around thirty fragments and is commonly known as The Vulgate, with the works based on it known as The Vulgate Tradition. These works consist primarily of that of Diodorus, the Bibliotheca historica, and Quintus Curtius Rufus, with his Historiae Alexandri Magni.
-- History of Alexander [The Vulgate/The Vulgate Tradition], by Wikipedia
If we compare the accounts of Alexander's settlement at Babylon, we find much the same information in Arrian and Curtius.29 Curtius, however, is much more interested in the citadel. Agathon, he says, was appointed citadel commander with a garrison of 700 Macedonians and 300 mercenaries, while the previous Persian commander, Bagophanes, was deprived of office but retained in Alexander's entourage.30 These arrangements are unrecorded by Arrian, but there is no reason to doubt them. The palace fortress to the north of Babylon had been the headquarters of the Persian garrison and dominated the city.31 Alexander must have placed his own forces in control, and the names and numbers in the vulgate tradition seem quite unexceptionable. At Babylon a Persian citadel commander was replaced by a Macedonian, and the same must have been true at Susa. Now Curtius reports the installation of a Macedonian (or Greek) named Xenophilus, who is attested in that role thirteen years later, whereas Arrian names an otherwise unknown hetairos with the Iranian name Mazarus. It is surely best to assume another misunderstanding on Arrian's part. His source presumably mentioned both the Persian commander, who can only have been Mazarus, and his Macedonian successor, Xenophilus. Arrian's narrative is compressed and superficial at this point, and, if he was working at speed, he might well have conflated the two men, so transforming Darius' garrison commander into a hetairos of Alexander.
If it is accepted that Mazarus held the citadel of Susa under Darius, a possible solution emerges for a famous numismatic [coins] problem. Barbarian imitations of Attic owl tetradrachms are a common phenomenon in the fourth century B.C.
An Athenian tetradrachm from after 499 BC, showing the head of Athena and the owl
The tetradrachm (Greek: τετράδραχμον, romanized: tetrádrachmon) was an Ancient Greek silver coin equivalent to four drachmae. In Athens it replaced the earlier "heraldic" type of didrachms and it was in wide circulation from c. 510 to c. 38 BC.
The Athenian tetradrachm was perhaps the most widely used coin in the Greek world before the time of Alexander the Great.
-- Tetradrachm, by Wikipedia
One large group, dating from the latter part of the century, presents particular problems.32 The provenance of these coins is invariably Babylonia, and they bear the Aramaic superscription of the striking authority, a name which Newell interpreted as 'Mazakes'. Similar imitations with the same superscription and the same peculiar monogram ([x] ) have been found in Egypt, and Newell rightly concluded that they were struck by Mazaces, the last Persian satrap of Egypt, in order to pay the Greek mercenaries levied before Issus.33 Less happily, he assumed that Mazaces was rewarded by Alexander for his surrender and given an autonomous territory in Babylonia, where he struck the imitation tetradrachms. Such an enclave of independence in Alexander's empire, with rights of coining included, is virtually inconceivable, and Badian quite justly dismisses the theory as 'a numismatists' myth'.34 The interpretation of the Aramaic superscription is by no means certain. Newell himself stresses 'the uncertainty inherent in so many Aramaic letters', and in the last century Six had read the name as 'Mazdad' or 'Mazdar', assuming that the coins were struck by Mazarus, whom he believed to have been Alexander's citadel commander in Susa.35 Once more the theory involved an anomaly. It is hard to envisage a Macedonian subordinate of Alexander issuing imitations of Athenian tetradrachms in his own name. But, once we assume that Mazarus held the citadel under Darius, the difficulties disappear. The barbarian imitations in Babylonia were issued at the same time as those from Egypt and with the same object, as payment from the vast number of Greek mercenaries in Persian service before Issus and Gaugamela. Mazarus, like Xenophilus in 317/16, presumably had the treasures of Susa in his charge,36 and he was the obvious person to supervise the striking of coins for the army. Such activity is certainly more plausible in the confusion of the latter years of Darius III than in the reign of Alexander, when the mints were rigidly controlled and a standard regal coinage enforced. If the legend on the Egyptian and Babylonian imitations is similar, it is explained by the similarity of the two names; both Mazaces as satrap of Egypt and Mazarus as citadel commander at Susa struck tetradrachms for the army and inscribed a virtually identical monogram.37
It should now be clear that it is unwise to insist on the precise wording of Arrian when that wording produces contradictions and conflict with other evidence. Arrian is prone to misread and misinterpret his primary sources, and the smooth flow of his narrative can obscure treacherous quicksands of error. There is a good example at 1. 6. 11, where Arrian claims that Alexander pursued his Illyrian opponents from the city of Pellion [x]. This can only mean one thing; the pursuit continued 'as far as the mountains of the Taulantians'.38 The Taulantians lived in the hinterland of Epidamnus/Dyrrhachium, and their territory probably extended as far inland as modern Elbasan.39 Wherever one locates the site of Pellion, the mountains of the Taulantians were some 100 kilometres from the battle site. A pursuit of those dimensions is, I think, impossible. Alexander launched his surprise attack on the Illyrian camp with a force of infantry, the Agrianians, hypaspists, and two phalanx battalions. He attacked before the rest of the army had caught up, and there is no hint of any cavalry support.40 What is more, only part of the enemy took to flight. A substantial group under King Cleitus withdrew to the city and prepared to stand siege.41 It seems incredible that Alexander should have divided his army and taken his infantry on a pursuit several days long. Once more it is easiest to assume a minor misunderstanding on Arrian's part. The narrative of the battle ends somewhat abruptly, for Arrian was no doubt keen to move on to the next major episode, the fall of Thebes, and he could have abbreviated his source misleadingly. Ptolemy need have said only that the pursuit was taken across the plain to the mountains which faced Taulantian territory (as opposed to those facing Macedonia). Arrian clumsily abbreviated his expression as 'the mountains of the Taulantians' and so created a pursuit of 100 kilometres.
II
So far the errors dealt with have been misunderstandings of a single source. There are, however, more complex problems created by faulty manipulation of two or more sources. At the outset Arrian states categorically that his two principal sources are Ptolemy and Aristobulus. Where they agree he will record the common version as the absolute truth; in case of disagreement he will select the more plausible version, making use of all memorable material.42 Arrian does not commit himself to record all divergences. Indeed he very rarely records disagreement, and gives variant traditions only when he considers them memorable in themselves. Usually he reproduces without comment the version which seems to him the more credible. There are accordingly two main areas where error is likely. Arrian may reproduce only one version, but he has read both, and there are occasional traces of contamination, both deliberate and inadvertent. The alternative, rejected tradition creeps in to infect the version chosen for reproduction. Secondly Arrian is not always aware when his sources are retailing the same episode. An incident may be placed by Ptolemy and Aristobulus at different points in the narrative or described with very different details. In these circumstances Arrian may use both descriptions from both sources and retail them as separate incidents. Both types of error can be traced, and I shall discuss some typical instances.
In his detailed description of the Macedonian battle line at Gaugamela Arrian reviews in order the six phalanx battalions and states that the battalion of Amyntas son of Philippus was commanded in his absence by his brother, Simmias.43 Amyntas' patronymic is clearly wrong. He and his brothers are mentioned repeatedly in the history of Alexander's reign, and their father's name is elsewhere unequivocally attested as Andromenes.44 There have accordingly been repeated attempts to emend away the offending name. Readers of Abicht's Teubner edition or Robson's Loeb edition will be unaware that Arrian's manuscripts read anything other than [x].45 But it is quite impossible to explain how the textual corruption arose. Nothing in the context suggests a reason for the scribe to have substituted [x] for [x]. The mistake can only be attributed to Arrian himself, and it is easy to see how it occurred. The vulgate tradition of Gaugamela includes a description of the Macedonian line. The phalanx commanders are the same as in Arrian, with one exception: the battalion of Amyntas was commanded in his absence by Philippus, son of Balacrus.46 Now we can explain the error in Arrian. His source for the Macedonian line of battle was most probably Ptolemy, and it was his version that Arrian followed. But he must have collated the Ptolemaic version against that of Aristobulus, and, although he accepted Ptolemy's statement that Simmias commanded Amyntas' battalion, he was aware of the variant tradition placing it under Philippus, son of Balacrus, and the name of Philippus slipped in erroneously as the patronymic of Amyntas. It seems to me a clear case of source contamination.
The same appears to have happened in Arrian's review of the Macedonian line at the Granicus (1. 14. 1-3). Here Arrian rather annoyingly lists the two halves of the army from the wing to the centre, giving first the right and then the left. The battalion of Philippus, son of Amyntas (a person otherwise obscure),47 occupied a central position in the line and is therefore mentioned at the end of both lists. Craterus' battalion, however, is also mentioned twice, but in different positions. In the review of the right half of the line it is placed between the battalions of Coenus and Amyntas, and it appears again as the battalion at the extreme left of the phalanx.48 Once more there have been attempts to delete one of the references to Craterus' battalion,49 but then it becomes impossible to explain the intrusion. There is no reason whatsoever to suppose a scribal gloss.
(1) Glosses. Ancient and medieval manuscripts contained many glosses, as defined by the usage of the word in Latin and not according to the original meaning of the word in Greek. In the study of ancient Greek and Latin texts the term ‘glossa’ carries a very distinct technical sense, which is also applied to medieval texts, though with some differences:2 [Oxford English Dictionary (2nd ed.; Oxford 1989) VI, 591. The Dictionary adds: ‘hence applied to a similar explanatory rendering of a word given in a glossary or dictionary. Also, in a wider sense, a comment, explanation, interpretation.’ The Oxford Classical Dictionary (2d ed.; Oxford 1970) [= OCD] subdivides the entry ‘glossa’ into two sub-entries, focusing on the meaning of the word in respectively Greek and Latin sources. For the former OCD provides the following definition: "In Greek literary criticism [x] meant any words or expressions (not being neologisms or metaphors) [x] (Arist. Poet. 1458b32), i.e. belonging not to the spoken language familiar to the critic (1458b6), but to a dialect, literary or vernacular, of another region or period (1457b4)." The modern use of the term ‘gloss’ does not reflect the meaning of that word in Greek, but rather that of the identical word in Latin, described as following in the same Dictionary: . . . marginal or interlinear interpretations of difficult or obsolete words.] ‘A word inserted between the lines or in the margin as an explanatory equivalent of a foreign or otherwise difficult word in the text.’3 [3 At a second stage these glosses were often collected, alphabetically or not, as so-called ‘glossae collectae’ or glossaries, and some of these actually constitute the basis of primitive dictionaries of equivalents. These glossaries were numerous in antiquity, and even more so in the Middle Ages. See especially the detailed description by B. Smalley, The Study of the Bible in the Middle Ages (3d ed.; Oxford 1983) 46-66. In fact, although the glossaries had their origin in the margins of manuscripts, once collected, they became independent units and thus started a life of their own.] Of a different nature are glosses in Sumerian and Akkadian texts (see below), since these glosses, often written within the text, were meant to be an integral part of that text.
-- Glosses, Interpolations, and Other Types of Scribal Additions in the Text of the Hebrew Bible, by Emanuel Toy
The most probable solution is again source contamination by Arrian. Both Ptolemy and Aristobulus presumably gave full descriptions of the battle line and differed over the position of Craterus' battalion. One placed it in the mass of the phalanx and the other at the extreme left, the position it was to occupy at Gaugamela and Issus.50 Arrian has absorbed both versions without reconciling the contradiction.
The account of the Granicus battle line is taken principally from Aristobulus. There are striking eccentricities when we compare it with other descriptions of major battles. The term used to refer to the Macedonian phalanx battalions is [x] not [x], the regular expression used in Arrian's narrative. This is a rare usage, not however confined to Arrian, and it seems to have been a technical term in the Hellenistic period.51 The terminology for the hypaspists is also obscure; they are called [x]. Elsewhere it is only the Macedonian cavalry who are termed Hetairoi and there is no indication that the class included infantry. Arrian may be making a mistake, misreading pezhetairoi, the generic title of the Macedonian infantry. Arrian's usage in other passages, however, suggests that pezhetairoi referred only to the six battalions of the phalanx, excluding the hypaspists.52 The terminology used for the Macedonian infantry at the Granicus is, to say the least, exceptional and indicates a source other than that regularly used in Arrian's battle descriptions, in other words, not Ptolemy but Aristobulus.
The terminology in the army list is not only unusual but varied. The most striking instance is the oscillation between [x] and [x] to refer to the cavalry under the command of Amyntas, son of Arrhabaeus. The same contingent is referred to in both instances, for its commander is the same and it is associated with the Paeonians and the ile of Socrates.53 The change of terminology is hardly due to Arrian himself, for the references to the cavalry of Amyntas are relatively widely spaced and there is no stylistic reason for variation. Both words are genuine Macedonian military terms,54 and there is no possibility of deliberate archaizing by Arrian. The only explanation seems to me that Ptolemy and Aristobulus used different terms; [x] is Aristobulus' term and [x] Ptolemy's. This oscillation helps to detect a doublet in the description of the preliminaries to the battle.
In etymology, two or more words in the same language are called doublets or etymological twins or twinlings (or possibly triplets, and so forth) when they have different phonological [speech sounds] forms but the same etymological [word origin] root. Often, but not always, the words entered the language through different routes. Given that the kinship between words that have the same root and the same meaning is fairly obvious, the term is mostly used to characterize pairs of words that have diverged at least somewhat in meaning. For example, English pyre and fire are doublets with merely associated meanings despite both descending ultimately from the same Proto-Indo-European (PIE) word *péh₂ur.
Words with similar meanings but subtle differences contribute to the richness of modern English, and many of these are doublets. A good example consists of the doublets frail and fragile. (These are both ultimately from the Latin adjective fragilis, but frail evolved naturally through its slowly changing forms in Old French and Middle English, whereas fragile is a learned borrowing directly from Latin in the 15th century.)
Another example of nearly synonymous doublets is aperture and overture (the commonality behind the meanings is "opening"). But doublets may develop divergent meanings, such as the opposite words host and guest, which come from the same PIE word *gʰóstis and already existed as a doublet in Latin hospes and then Old French, before being borrowed into English. Doublets also vary with respect to how far their forms have diverged. For example, the connection between levy and levee is easy to guess, whereas the connection between sovereign and soprano, or grammar and glamour, is harder to guess.
-- Doublet (linguistics), by Wikipedia
In the course of the approach to the Granicus Alexander sent out a party of scouts under the command of Amyntas, son of Arrhabaeus; they comprised four ilai of [x] and the ile of Socrates (1. 12. 7). After a short passage describing the Persian council-of-war at Zeleia Arrian reverts to the Macedonian army. Alexander led his army in double column and sent scouts ahead under the command of Hegelochus, comprising the [x] and 500 light infantry (1. 13. 1). It is possible that two groups of scouts were sent out, Hegelochus commanding the remnants of the [x] not included in Amyntas' command.55 There is no suggestion from Arrian's wording that Hegelochus' men formed a second group, and it is striking that his terminology for the cavalry has changed. Arrian refers to the first group as [x] and the second as [x]. He has changed sources in the interim. He used Aristobulus' account of the Persian council-of-war (the names of the leaders given there differ slightly from those in the later casualty list),56 and continued with Aristobulus for the debate between Alexander and Parmenion and for the battle order of the Macedonian army. The narrative switches back to Ptolemy for the actual attack (1. 14. 5). The two references to the scouts are taken from different sources, and it is possible that Ptolemy and Aristobulus disagreed over the composition and leadership. Arrian has failed to reconcile the divergence and so reported the mission of the scouting party twice over.
There are even more striking examples of Arrian's maladroit use of sources in the narrative of Alexander's Indus voyage in 325. At 6. 15. 5, just before his account of the invasion of the land of Musicanus, Arrian states that Craterus was sent out again with the army through Arachosia and Drangiana: [x]. As it stands, the text refers to Craterus' commission to escort part of the infantry and the disabled veterans down the Helmand valley to Carmania. Unfortunately Arrian describes that commission very explicitly two chapters later (6. 17. 3), and in the intervening narrative he says that Craterus was given the task of fortifying the citadel of Musicanus' capital (6. 15. 7). It is impossible that Craterus could have been sent twice on his way, and Schmieder accordingly deleted [x] as a scribal gloss. Craterus, he assumed, was ordered to accompany the fleet along the Indus bank, as he had done in the earlier stage of the journey (6. 15. 4). Schmieder's solution is attractively simple, and it has been universally accepted.57 It is, however, untenable. In the first place, when the excision is made, Arrian's text does not have the meaning required by Schmieder. His language still implies that Craterus was sent away from Alexander and did not accompany the fleet. The verb [x] is invariably used by Arrian to denote sending a person or contingent either temporarily or permanently from Alexander's court.58 The closest parallel I can find for the sense required by Schmieder is a passage at the beginning of the Indus journey (6. 4. 1-2). Here Alexander sends out ([x]) two infantry columns to meet his fleet near the confluence of the Acesines and Hydraotes (cf. 6. 5. 5). The land forces go to the same destination but by a different route. That might have been true of Craterus' commission at 6. 15. 5, but we should still expect some qualification of destination or purpose. Schmieder's deletion leaves the text puzzlingly elliptical; Craterus is sent out, we know not where.
More seriously, the gloss posited by Schmieder is extremely sophisticated.59 We have to assume that the scribe was aware of Craterus' mission to Carmania from his reading of 6. 17. 3 and inserted an erroneous note that the earlier mission of Craterus was through Drangiana and Arachosia, a note that was later absorbed into the text. It is a doubly sophisticated procedure in that the wording is changed. At 6. 17. 3 Arrian refers to the route as [x], whereas in the supposed gloss the Zarangae are called by their synonym, [x]. It is surely unlikely in the extreme that a reader copying the route of Craterus from the later passage would have substituted the correct synonym for a people so obscure.60 It is much easier to suppose that Arrian has included a doublet of Craterus' mission, placed by his two sources at different stages of the voyage down the Indus. Of these sources one referred to the [x] and the other to the [x]. We can detect the same oscillation of nomenclature in the earlier part of the narrative, which deals with the march of Alexander through Drangiana, the area around the Helmand Lakes in modern Sistan. This was the old Persian satrapy called Zranka or 'sea land'.61 At 3. 25. 8 Alexander is said to have reached the palace of the [x], while in the resumptive note at 3. 28. 1 he refers to the people as [x].62 In the latter passage Arrian has patently changed sources, for after describing the settlement of the Ariaspae, the southern neighbours of the Drangae located on the Helmand river by the Arachosian border, Arrian gives a brief resume of the journey from Prophthasia, including the submission of the Drangae, whose territory Alexander had already traversed. The variation in nomenclature is hardly due to Arrian himself; it is much more likely to be the result of divergence in his primary sources. Perhaps Aristobulus used the form [x] which appears to be the more common Hellenistic form,63 and Ptolemy used the synonym [x], the older form found in Herodotus, which is an exact translitteration of the Akkadian.64 The same variation between 6. 15. 5 and 6. 17. 3 is confirmation that the two passages are derived from different sources.
Ptolemy and Aristobulus placed the mission of Craterus at different points in the narrative. One located the starting-point near the southern tip of the Punjab and the other towards the Indus delta, in the vicinity of Patala. Both points were salient for the Bolan Pass, the principal route through the mountains into Arachosia and certainly the route taken by Craterus. There is one consideration which tells in favour of the northern location. If Craterus had been sent back from Patala he would have needed to retrace his steps up the Indus before branching off to the pass, whereas, if he had left Alexander immediately south of the Punjab, he merely needed to diverge westwards. For what he is worth, Justin supports the northern starting-point. The army column, he says, was sent to Babylonia between the territory of the Malli and the kingdom of Sambos;65 that corresponds roughly to the first location in Arrian. Even if wrong, Justin confirms that there was an alternative tradition about the starting-point of Craterus' march, a tradition which appears in the first part of the doublet in Arrian and also in Strabo's brief account of the return to the west.65a That a doublet exists can hardly now be doubted. Schmieder's sarcastic question, 'who would believe that Arrian after so few words forgot what he had just written?', is easily answered. The man who could place Craterus' battalion in different positions of the battle line, in consecutive paragraphs of his narrative of the Granicus, was perfectly capable of an inept doublet of Craterus' mission through Arachosia.
Immediately before the first reference to Craterus' march there is another error. This occurs in the description of the establishment of the satrapy of Southern India. As satrap of the territory between the Ocean and the confluence of the Indus and Acesines Alexander appointed Oxyartes and Peithon.66 Two satraps in a single province are a unique phenomenon under Alexander, and Kruger conjectured that Peithon was merely a general, the assistant and supervisor of a native satrap. That will not do. Peithon is elsewhere attested satrap of Southern India (6. 17. 1), and there is no hint that he had a colleague. What is more, Oxyartes is undoubtedly the father-in-law of Alexander. He is mentioned by name in the previous paragraph as the newly appointed satrap of Parapamisadae, the district around the Kabul valley. The appointment is mentioned by the vulgate as well as by Arrian.67 Again the most popular course has been to delete Oxyartes' name from the text at 6. 15. 4,68 but there is no reason for any scribe to have glossed Peithon's name with that of Alexander's father-in-law. The mistake is certainly Arrian's own, and there are two possible explanations. The first is simple misunderstanding. Both Oxyartes and Peithon retained their satrapies after the distributions at Babylon and Triparadeisus. Both appear in the satrapy lists, and the nomenclature of Peithon's satrapy is interesting; it comprised the parts of India contiguous with Parapamisadae, Oxyartes' satrapy ([x]).69 Indeed Oxyartes and Peithon are listed in sequence. Now the satraps and satrapal boundaries confirmed for the east at Babylon were those established by Alexander himself.70 Arrian's source may have specified that Alexander fixed a common boundary for Parapamisadae and Southern India, and that Oxyartes and Peithon were satraps of the respective territories. Arrian mentioned only Southern India, but he had Oxyartes' name on his mind and added it inadvertently to the narrative.
There is another possibility, that the entire paragraph (6. 15. 4) is a compressed doublet. Arrian's account of the journey from the Malli town to the kingdom of Musicanus is extremely difficult to reconcile with the common tradition of Curtius and Diodorus, explicitly derived from Cleitarchus.71 In the Cleitarchean version Alexander first travels with the fleet, Craterus leading the shore army. He receives the submission of the democratically governed Sambastae and moves to the next peoples, whom Diodorus terms the Sodrae and Massani.72 These peoples also submit, and Alexander ceremonially founds a city in their territory. He then installs Oxyartes in Parapamisadae. The next event is the invasion of Musicanus' kingdom. In Arrian's parallel account there is a brief note about the departure of the fleet from the base camp at the junction of the Hydraotes and Acesines (6. 14. 4). After a geographical excursus on the rivers of the Punjab he mentions Alexander's next halt at the confluence of the Acesines and the Indus. Here he awaits the arrival of Perdiccas' army group, which had subjugated the Abastani (6. 15. 1). This is Arrian's only reference to this mission of Perdiccas, but the Abastani are presumably the Sambastae who are the first to be mentioned in the vulgate version of the journey south.73 Perdiccas may have dealt with the tribes away from the river while Alexander received the submission of the people in the vicinity of the river.74 According to Arrian the wait at the confluence was protracted. While Alexander was there, additions arrived for the fleet and he received the submission of various autonomous tribes including the [x] who may be identical with the Massani of Diodorus.75 Still at the confluence he appointed Philippus satrap of Northern India and gave instructions for the foundation of a large city complete with dockyards. This is exactly the point at which Diodorus and Curtius attest the foundation of an Alexandria.76 Arrian, it is true, says that Alexander merely gave orders for the foundation, but he was on the site of the future city and presumably began the work of foundation as he had done at Alexandria Eschate.77 Finally, as in Cleitarchus' version, Alexander appointed his father-in-law satrap of Parapamisadae. At this point we expect the invasion of Musicanus' kingdom. Instead Arrian seems to revert to an earlier stage of the journey. Craterus and the land army march alongside the river, as they do in the vulgate account of the departure from Malli territory.78 Next comes a reference to the arrival at the palace of the Sogdi, who look like the Sodrae of Diodorus. Here too Alexander founds a city, again with dockyards, and again there is a reference to the fleet; the ships which had been damaged were refitted. Why, one asks, had that not been done at the confluence, where Alexander had deliberately lingered to concentrate his forces? Why also should a second Alexandria have been founded in such close proximity to the city at the confluence? There is no hint elsewhere of two foundations in this area and no trace of either city in later history.79 The difficulties evaporate if we assume a doublet in Arrian. His sources will have given different accounts of the journey from the camp near the Malli town to the confluence of the Indus and Acesines, recounting the march along the bank of the river and the foundation of an Alexandria but with different details and different forms of the Indian names. Arrian accordingly gave both versions consecutively as separate episodes. We cannot assume too long a distance between the territories of the Malli and Musicanus. In a survey of the tribes of Southern India, probably derived from the contemporary Onesicritus, Strabo places the kingdom of Musicanus immediately downstream from the land of the Malli;80 unless there is a doublet in Arrian's narrative the interval between the two districts becomes uncomfortably large.
If there is a doublet in Arrian, we should expect the doublet to end with a reference to satrapal appointments, and in particular a reference to Oxyartes' installation in Parapamisadae. If such a reference occurred it would explain very nicely the confusion about the double satrapy. Arrian had already mentioned the appointment in Parapamisadae and felt no need to repeat it. He therefore mentioned only Peithon's appointment to Southern India, but the name of Oxyartes was clearly in his mind and it slipped in inadvertently. The confusion would be all the easier if his source mentioned the common boundary of the two satrapies. It is an error parallel to the substitution of Philippus for Andromenes as the patronymic of Amyntas. This section of the narrative of the Indus voyage is an outstanding example of maladroit conflation of primary sources. The errors are blatant and striking, and it is useless to attempt to restore Arrian's credit by the crude surgery of deleting the offending words.