1. Home
  2. Antiquity
  3. Conquest of the Achaemenid Empire
  4. Granicus

Battle

Granicus

May 334 BC

Cross under pressure, widen the fight, isolate the infantry

Military history

How did a contested crossing widen enough to break the Achaemenid cavalry and then isolate its mercenaries?

At the Granicus, Alexander faces not Darius, who is still absent, but the forces of the Anatolian satraps. The tradition preserved by Arrian and Plutarch places the Achaemenid cavalry on the eastern bank: an initial Macedonian wave struggles to leave the riverbed, then the main attack widens the melee until the mounted line gives way. Diodorus gives an incompatible opening, with an unopposed crossing at dawn before deployment. The map chooses the first sequence to make the mechanism legible; it does not turn either that narrative choice or the modern hypothesis of a calculated feint into certainty. Victory opens the roads of western Anatolia, but destroys neither Darius’s royal army nor the Achaemenid Empire’s capacity to resist.

A royal army still on the march

Alexander has just crossed into Asia with an army that the narratives count at campaign scale, not only as the force on this battlefield. Darius III has not yet concentrated the royal army: the Anatolian satraps and Memnon of Rhodes must decide how to stop the invasion with their regional forces.

Every option at the council of Zeleia carries a cost

Arrian and Diodorus report that Memnon advises destroying local resources and avoiding battle. Arsites refuses to ravage his satrapy; other Persians also suspect the Greek of pursuing his own interest. This debate comes through a Greek tradition and is not a set of council minutes, but it captures a real dilemma: surrender land and revenue, or risk a regional army to block the road at once.

One river, two battles in the sources

In Arrian, Parmenion considers a crossing too risky as evening approaches: current, depth, and banks could disorder units under missile fire. Alexander attacks nevertheless, facing cavalry massed on the opposite bank. Plutarch preserves the same general difficulty. Diodorus has both armies camp and then the Macedonians cross unopposed at dawn before forming their line. These openings cannot be combined into a single chronology.

The reach and limits of the first wave

Arrian names an advance force of cavalry, Paeonians, and infantry that enters the river first, receives missile fire, and falls back under pressure. Its movement draws or fixes part of the Persian left before Alexander arrives. Badian and Devine read this as a maneuver intended to disrupt the enemy; that is a plausible reconstruction, not proof of a planned sacrifice.

Military history

Sources

Heritage archive edition · late ancient chronicle based on campaign traditionsArrien, Anabase d’Alexandre I.11–17, traduction E. J. Chinnock

I.11.3 : plus de 30 000 fantassins et plus de 5 000 cavaliers ; I.12.6–10 : conseil de Zélée et proposition attribuée à Memnon ; I.13.1–7 : approche, renseignement, avis de Parménion et décision d’attaquer ; I.14.1–7 : ordres de bataille et première vague ; I.15.1–8 : difficulté de la sortie, engagement oblique et mêlée ; I.16.1–7 : rupture, poursuite limitée, mercenaires, pertes et prisonniers ; I.17.1–3 : Sardes.

Structure of the map and main narrative, force composition, decisions, movements, the separate fight against the mercenaries, casualties, and immediate consequences.

View ↗
Late ancient chronicle · alternative traditionDiodore de Sicile, Bibliothèque historique XVII.17–21, traduction C. Bradford Welles

XVII.17.3–5 : catalogue de 32 000 fantassins ; le total déclaré de 4 500 cavaliers contredit ses postes, qui en additionnent 5 100 ; XVII.18.2–19.6 : conseil perse, effectifs achéménides et franchissement à l’aube sans opposition ; XVII.20.1–21.4 : combat de cavalerie et rupture ; XVII.21.5–7 : pertes, prisonniers et reddition de Sardes.

The incompatible opening, a second force catalog, another account of the fighting, and a divergent casualty range.

View ↗
Late ancient biographyPlutarque, Vie d’Alexandre 16, traduction Bernadotte Perrin

§16.1–3 : passage difficile, courant, boue et treize escadrons de cavalerie ; §16.4–6 : mêlée, combat personnel et intervention de Cleitos ; §16.7–8 : phalange, demande de quartier des mercenaires et traditions de pertes.

A parallel biographical tradition on the defended crossing, close combat, the mercenaries’ position, and the alternative death toll; it is not treated as independent contemporary testimony.

View ↗
Specialist academic studyErnst Badian, « Granicus », Encyclopaedia Iranica, vol. XI, fasc. 2, p. 187

PDF p. 1 / p. 187 : fin mai, identification au Kocabaş Çayı, basses hauteurs de 3–4 m, effectif perse irrécupérable, berges raides alternant avec des plages de gravier et première vague reconstruite comme appât ; PDF p. 2 : mercenaires, Arsitès et portée immédiate.

Topographical checks, criticism of the reported strengths, a modern reconstruction of the first crossing presented as a hypothesis, and the limits of reconstruction.

View ↗
Academic synthesis of the Achaemenid EmpirePierre Briant, « Alexander the Great », Encyclopaedia Iranica, vol. I, fasc. 8, p. 827–830

HTML, p. 827, paragraphes 3–5 : débarquement de 334, armée royale pas encore pleinement mobilisée, victoire du Granique et limites de la soumission réelle en Anatolie ; section « Resistance to Alexander », pp. 829–830 : défaite militaire sans effondrement immédiat et résistance à Halicarnasse.

Achaemenid context, Darius’s absence from the Granicus, regional consequences, and rejection of a teleological reading of its strategic significance.

View ↗
Specialist academic articleA. M. Devine, « Demythologizing the Battle of the Granicus », Phoenix 40.3, 1986, p. 265–278

p. 276 : confrontation des pertes macédoniennes d’Arrien et de Plutarque, total d’environ 115 chez Arrien ; p. 277 : la première vague interprétée explicitement comme une « feint » et critique de l’amplification héroïque et stratégique du Granique.

Criticism of casualty figures and heroic framing; the modern first-wave-decoy hypothesis retained conditionally and never converted into an attested intention.

View ↗
University archaeological and geomorphological surveyReyhan Körpe, « 2022–2024 Granicus Battlefield Survey: New Findings, New Results », Troy Museum Journal, 2025, p. 107–123

PDF p. 1 / revue p. 107–108 : hypothèse d’un champ plus méridional et de Höyükbayırı Tepe ; PDF p. 6 / revue p. 113 : plaine classique marécageuse au IVe siècle, évolution alluviale et absence de mobilier de bataille ; PDF p. 14–15 / revue p. 121–122 : résultats encore provisoires et poursuite des sondages.

Updating discussion of the site, paleochannel, and final rise without converting a survey proposal into a certain location.

View ↗
Specialist academic articleN. G. L. Hammond, « The Battle of the Granicus River », Journal of Hellenic Studies 100, 1980, p. 73–88

p. 73, résumé et ouverture : divergence des sources sur l’heure et le but du passage, problème du mouvement latéral et inadéquation partielle entre topographie ancienne et terrain moderne ; n. 34 : contradiction entre le total déclaré de 4 500 cavaliers chez Diodore et les 5 100 obtenus par addition.

Identifying chronological, oblique-advance, and topographical problems; arithmetic check of Diodorus’s catalog, without imposing the article’s cartographic solution.

View ↗

Edited by BattlesInMotion.com