In 522 BC Cambyses, king of Persia, died after hearing a report that a magus, impersonating his younger brother, had usurped the throne. Several months later Darius killed the usurper and ascended the throne, becoming king in his place (Herod. 3:61-88). Such is the account of events handed down to us by the ancients in the form of a rock inscription at Behistun, commissioned by Darius himself, and a section of Herodotus thirty chapters in length. The intent of this paper is to assess the reliability of the Herodotean account of the accession of Darius and, as such, several factors will need to be addressed. The most important will necessarily be the nature of his sources. However, also of importance is Herodotus’ thematic concern of hubris and divine retribution. In an address of such factors an analysis of Herodotus’ tendency towards both the use of anecdotes and the evocation of the supernatural must necessarily assume a prominent position.

The core of Herodotus’ account of the accession of Darius finds its source in the official version established by Darius himself.[1] ‘The Histories’ and the Behistun inscription agree that: Cambyses killed his brother, Smerdis (Herod. 3:30; BI. 1.26-35); that he died on his return to Persia (Herod 3:64-66; BI. 1.35-43); that a magus usurped the throne, pretending to be Smerdis (Herod. 3:61; BI. 1.35-43); and that Darius defeated the usurper with six co-conspirators (Herod 3:71-78; BI. 1.48-61; BI. 4.80-86). both accounts also agree on six of the names of the seven co-conspirators.[2] However, those details which Herodotus and the Behistun inscription have in common amount to a negligible percentage of Herodotus’ entire account. Furthermore, the reliability of those details which Herodotus shares with the official account is not settled simply in point of that fact, as the reliability of the Behistun inscription is, itself, called into question by its propagandistic nature.[3]  Nevertheless, it must be acknowledged that Herodotus received his information from those copies which Darius had distributed throughout the empire (BI 4.88-92), rather than the Behistun inscription which was unreadable from ground level.[4] As such, we cannot truly know exactly how much of Herodotus’ account has its source in some form of the official tradition. Yet, even if the most liberal estimate is given, it must be accepted that Herodotus has left the historian with no means of ascertaining the sources to much of his information pertaining to the issue at hand.

Whatever Herodotus’ sources may have been, it is likely that they were of an oral, rather than a written, nature. While it is mentioned in the Behistun inscription that Darius had his official account distributed throughout the empire (BI. 4.88-92), whether there existed a Greek translation at the time of the compilation of ‘The Histories’ is unknown. Nevertheless, a heavy reliance on oral sources does not necessarily lead to a complete misrepresentation of events. However, in the case of the Herodotean account of the usurpation of the Persian throne a lack of understanding of the Persian language and a subsequent reliance on oral sources has led to an error of sizeable importance. According to the Behistun inscription, one magus, named Gaumata, usurped the Persian throne, pretending to be Cambyses brother, Smerdis (BI. 35-43). Such an account is in stark contrast with Herodotus who records that, not only was the magus actually named Smerdis, but that he was assisted by his brother Patizeithes (Herod 3:61). The name Patizeithes does not appear in Persian sources[5] and it has been argued that what has occurred is that Herodotus has created a second character through a misunderstanding of a Persian word roughly meaning ‘viceroy’, most likely the title of whomever Cambyses left as regent while in Egypt.[6]

Furthermore, the reliability of Herodotus is brought into question simply by the presence of the claim that the throne was usurped by a magus. It is on the issues of the murder of Cambyses’ brother and the usurpation of the throne that Herodotus most closely follows the official account as presented in the Behistun inscription. However, it is these sections of the official account that modern scholars treat with the most suspicion.[7] The timing of the murder of Smerdis could not be agreed upon even by the ancient authors: Darius places it before the Egyptian expedition (BI. 1.26-35), Herodotus during (Herod. 3:30), Ctesias after.[8] Herodotus also admits that the manner of the prince’s death was not known with any certainty, recording two widely varying accounts of his assassination (Herod. 3:30). It is no surprise, therefore, that many historians doubt whether the murder of Smerdis was anything more than a fiction formulated by Darius to conceal the fact that he had stolen the throne from the son of Cyrus.[9] If it is accepted that Smerdis was the one whom Cambyses left as regent in his absence and that he seized the throne then the difficulties of how the death of the heir apparent was kept secret for some four years and how a pretender managed to fool the entire royal household into believing that he was their relative disappear. While the sources do not provide us with enough information for any statement on the true personhood of the usurper to be anything more than informed speculation, it is likely that it was Cambyses’ younger brother, rather than a magus. This being the case, the reliability of the Herodotean account has reason to be doubted as, at its core, it may be a retelling of political fiction.

Were the hypothesis above, that Cambyses’ younger brother was the one who usurped the throne, to be disproved, Herodotus’ reliability on the issue of Darius’ accession would still be suspect as a result of the presence of a large number of anecdotes and details which prove to be fictitious. At least twice in his account Herodotus records the speech and actions of personalities for which he could not have had a source. Given that the existence of the character of Patizeithes has been established to be the result of linguistic confusion, the conversation between the two magi and Prexaspes could never have taken place (Herod 3:74-75). Similarly, the exchange between Darius and Gobryas as they assassinated the magus would not have been information which Herodotus could possibly have attained (Herod 3:78). Furthermore, Marincola argues that the account of Phaidime feeling for the king’s ears betrays a Greek tradition as the idea that a Persian king’s ears were usually covered finds its source in Greek art.[10] Though perhaps the most damning error on the part of Herodotus is his placing of the deposition of the king at the palace in Persia, rather than the fortress in Media mentioned in the Behistun inscription (Herod. 3: 76-78; BI. 1.48-61).

Herodotus’ reliability is further threatened as he strays into the realm of the supernatural in order to explain events. It is likely no accident that Herodotus bookends his account of the accession of Darius with examples of hubris, the main theme flowing through his work. The Herodotean account opens with the recurring motif of the hubristic king, in this case Cambyses, receiving an oracle forewarning his fate and subsequently attempting to overcome it only to find that his own efforts are what bring about his demise (Herod. 3:30; Herod 3:63-64). Herodotus also embellishes the official account of Cambyses death so as to continue his theme of divine retribution. While the accurate translation of the Behistun inscription on the matter of Cambyses’ death has been disputed amongst scholars, some arguing natural causes, others suicide[11], whichever proves true after further study, neither will support the details of Herodotus’ account which relate a thigh wound in the exact place of the slaughter of Apis (Herod. 3:64). The Herodotean account closes with the recount of Darius’ deception in the hippomancy used as the tool to decide kingship (Herod 3:84-87). This deception could equally be referred to as a usurpation of the power of Ahuramazda as the god’s animal, the horse, is being coaxed into giving Darius’ decision rather than Ahuramazda’s. The presence of divine retribution for hubris in the account of Cambyses may, therefore, be intended to foreshadow the inescapable retribution for Darius’ hubris which will manifest itself in the form of the failed invasion of Greece. Such a literary, thematic construction in Herodotus’ account of Darius’ accession severely damages its reliability.

One of the areas of greatest contention amongst historians in regards to the reliability of the Herodotean account is that of the political discussion between the seven conspirators. Herodotus records that, subsequent to the deposition of the magus, the conspirators discussed the merits of various political systems in order to choose that which would best suit the Persian empire (3:80-83). While some simply discount this discussion as absurd, if Dandamaev’s argument that the conspiracy against the king was a result of the tribal elite attempting to regain the powers and rights that had been taken from them it would follow that some political discussion would have taken place, even if the form presented in Herodotus is fictional.[12] As Brannan notes, the main argument provided against the historicity of this discussion is that Herodotus records Otanes as supporting the concept of democracy and that such a system was incompatible with the oriental mind.[13] However, Brannan argues that such an understanding of Otanes speech is misinformed. The word ‘democracy’ never appears in Otanes’ speech, whereas concepts such as ‘equality under the law’ do.[14] Dandamaev agrees with Brannan on this issue and concludes that the political system which Otanes is supporting is not an Athenian style democracy but rather a tribal form in which the tribal nobility hold sway.[15] Such interpretations can be nothing more than informed speculation, however their logic and consistency are strong enough to conclude that the discussion in Herodotus cannot reasonably be disregarded out of hand.

A similar level of uncertainty surrounds the manner in which Darius became king. As mentioned above, Herodotus records that it was decided amongst the conspirators that hippomancy would be used and that the owner of the first horse to neigh at sunrise would become king (Herod. 3.84-87). Dandamaev notes that all classical authors agree on this account and accepts the report as historical without question, citing as further evidence another passage in Herodotus where, he argues, it is mentioned that the king of Persia is usually appointed by a portent, or by the decision of the people.[16] However, such a conclusion is the result of a misunderstanding of the text: Otanes relates that they will have to employ such methods to choose the king in this instance (Herod. 3.83), but that is not enough to conclude that such is the way it has always been done. Briant argues the opposite way and claims that the account of hippomancy is fictional and was formulated after Darius became king as a means of propagating his claim to have been chosen by Ahuramazda.[17] However, the Behistun inscription plays down the importance of the other six co-conspirators in order to focus on the exploits of Darius with an eye to establishing his right to rule (BI. 1.48-61). For Darius to later create a new story which put him on the same level as his co-conspirators would be counter-productive. Furthermore, the existence of the accounts of Darius cheating which Herodotus records (Herod. 3.85-87) betrays that the motif of hippomancy may have been invented, not by those seeking to establish Darius’ reign, but by those seeking to undermine it. It must, therefore, be concluded that the historicity of the hippomancy recorded in Herodotus is unknown.

However, the presence of the political discussion and the hippomancy in the Herodotean account, reliable or not, demonstrates that Herodotus correctly understood the pedigree of Darius. In the Behistun inscription Darius presents himself as of royal descent (BI. 1.1-11) and, therefore, as the rightful heir to the throne once the usurper had been deposed. However, Herodotus does not recognise such descent in his account and not once does he imply that Darius had any more right to the throne than his co-conspirators. In fact, Herodotus himself states that Darius was originally a man of ‘no great importance’ (Herod. 3.139), a description which could not have been applied to royalty. On this point scholars now concur with universal assent, making clear that the genealogy provided in the Behistun inscription was a fabrication.[18] As Dandamaev states, were Darius of royal descent then it would have been his father, Hystaspes, who ought to have been crowned king.[19] The reliability of Herodotus regarding the pedigree of Darius likely points to a reliable source from which Herodotus drew at least some of his information. However, as argued at the commencement of this paper, the fact that Herodotus has left the historian with few means of discerning his sources means that it likely cannot be deduced whether other aspects of his information come from said reliable source. As such, the reliability of this one aspect of the Herodotean account of the accession of Darius cannot be used to argue for its reliability as a whole.

The reliability of the Herodotean account of the accession of Darius is, in many respects, lacking. Herodotus had access to the official account in some form and used it to create the core of his account. However, the official account has proven to be propagandistic and most unreliable on the issues in which Herodotus has put the most faith. As such, the core of the Herodotean account, apart from the assassination of the king by Darius, likely has no basis in historical fact. Onto this unsubstantiated core, Herodotus has attached numerous fictitious anecdotes and appeals to the supernatural which add nothing but drama to the account and which serve to promulgate his recurring moral motif of hubris and divine retribution. However, Herodotus has demonstrated that he had access to a reliable source which related to him the true pedigree of Darius. This same source may have been the origin for the Herodotean accounts of the hippomancy and the political discussion, however this cannot be ascertained. Ultimately, the Herodotean account is unreliable and gives no further insight into how Darius ascended the throne than what can be gathered from the Behistun inscription, unless highly tendentious speculation is employed.



Ancient Sources

Avesta, (n.d.). Old Persian Texts. Retrieved from: (accessed May 27 2017)

Herodotus, The Histories, trans. Aubrey de Sélincourt. Introduction and notes John Marincola (London: Penguin, 2003).

Modern Sources

Bickerman, E.J. (1978). ‘Darius I, Pseudo-Smerdis, and the Magi’, Athenaeum, vol. 56, pp. 239-261

Brannan, P.T. (1963). ‘Herodotus and History: The Constitutional Debate Preceding Darius’ Accession’, tradition, pp.427-438

Briant, P. (2002). From Cyrus to Alexander: A History of the Persian Empire. Trans. Peter T. Daniels. (Winona Lake, Indiana:Eisenbrauns).

Dandamaev, M.A. (1989). A Political History of the Achaemenid Empire. (Leiden, the Netherlands: E. J. Brill).

Lang, M.L. (1992). ‘Prexaspes and Usurper Smerdis’, Journal of Near Eastern Studies, vol. 51, no. 3, pp. 201-207

Olmstead, A.T. (1938). ‘Darius and his Behistun Inscription’, The American Journal of Semitic Languages and Literatures, vol. 55, no. 4, pp. 392-416

Shayegan, M.R. (2006). ‘Bardiya and Gaumata: An Achaemenid Enigma Reconsidered’, Bulletin of the Asia Institute, vol. 20, pp. 65-76





Livius. (2017). ‘The Behistun Inscription’. Retrieved from: (Accessed May 21 2017)

Livius. (2015). ‘Cambyses II (2)’. Retrieved from: (Accessed May 26 2017)

[1] Bickerman, E.J. (1978). ‘Darius I, Pseudo-Smerdis, and the Magi’, p.239; Lang, M.L. (1992). ‘Prexaspes and Usurper Smerdis’, p.202; Olmstead, A.T. (1938). ‘Darius and his Behistun Inscription’, p.395

[2] Briant, P. (2002). From Cyrus to Alexander: A History of the Persian Empire, p.108

[3] Bickerman, E.J. (1978). p.239

[4] Livius. (2017). ‘The Behistun Inscription’. Retrieved from:

[5] Herodotus, The Histories, (London: Penguin, 2003), p.647

[6] Shayegan, M.R. (2006). ‘Bardiya and Gaumata: An Achaemenid Enigma Reconsidered’, p.66-67

[7] Bickerman, E.J. (1978). p.246-247

[8] Olmstead, A.T. (1938). p.396

[9] Dandamaev, M.A. (1989). A Political History of the Achaemenid Empire, p.91; Lang, M.L. (1992). p.201

[10] Herodotus, The Histories, (London: Penguin, 2003), p.648

[11] Livius. (2015). ‘Cambyses II (2)’. Retrieved from:

[12] Dandamaev, M.A. (1989). p.106

[13] Brannan, P.T. (1963). ‘Herodotus and History: The Constitutional Debate Preceding Darius’ Accession’, p.431

[14] Ibid. p.432-433

[15] Dandamaev, M.A. (1989). p.107

[16] Ibid. p.107

[17] Briant, P. (2002). p.111

[18] Brannan, P.T. (1989). p.431; Briant, P. (2002). p.111; Dandamaev, M.A. (1989). p.107

[19] Dandamaev, M.A. (1989). p.107