Abstract
The argument that Late Bronze Age Greece lacked a collective political name, and possibly lacked a hegemonic state, rests substantially on absence: the Linear B archives preserve no ethnonym for the Greeks, no correspondence between palaces, and no vocabulary of superordinate kingship. Arguments from absence are legitimate but weak, and they are also self-reinforcing, since a discipline that has decided what is missing tends to under-recognize it when it arrives. This paper attempts the reverse operation. It specifies, in advance of discovery, what documentary evidence would have to be found to support the maximalist reading — that Mycenae or another center exercised named authority over other Aegean kingdoms — and what would have to be found to establish that Bronze Age Greeks understood themselves as one people among the peoples of the world. It grades hypothetical finds by decisiveness, identifies the false positives most likely to be over-read, considers where such documents could physically survive, and works through four counterfactual discovery scenarios, including a null scenario in which further finds strengthen the minimalist case. The purpose is to fix evidentiary standards before the spade turns rather than after.
1. Why the Question Is Live
This is not an idle exercise. The Aegean documentary corpus is growing. The Odos Pelopidou excavations at Thebes produced a body of Linear B tablets published from 2001 onward, altering the picture of Boeotian palatial administration and religion (Aravantinos, Godart, & Sacconi, 2001). A single inscribed tablet fragment recovered at Iklaina in 2010 demonstrated that writing was in use at a second-order center within the Pylian kingdom, not merely at the palace, with implications for how administrative authority was distributed (Cosmopoulos, 2019). Excavations at Ayios Vasileios in Laconia have since 2008 produced the first palatial archive from a region that Linear B had previously left blank.¹ The Griffin Warrior burial at Pylos, opened in 2015, transformed assumptions about elite display in the transitional period (Davis & Stocker, 2016). Every decade of the last half-century has added material that no one had predicted.
The question, therefore, is not whether new texts will appear but whether the field has agreed in advance what they would have to say. Absent such agreement, three failure modes are likely. First, ambiguous finds will be read into whichever framework the reader already holds. Second, genuinely decisive finds may be discounted because they are singular. Third, and most insidiously, a slow accumulation of suggestive items may be treated as cumulative proof when each item is separately explicable by other means.
2. Four Claims, Not One
Any evidentiary program must begin by separating claims that ordinary discussion runs together. The following are logically independent:
(1) Hegemony existed. Some Aegean ruler exercised effective power over rulers in other regions — the ability to require deliveries, levy troops, or determine succession.
(2) Hegemony was named. That power had a title, a vocabulary, and a documented protocol distinguishing superior from subordinate.
(3) Hegemony was acknowledged from below. Subordinate administrations recorded their own obligations in those terms, rather than the superior alone asserting them.
(4) Corporate identity existed. Aegean Greek speakers classified themselves as a single category distinguishable from other peoples, whether or not any political structure united them.
Claim 4 is separable from 1–3 in both directions. Classical Hellas demonstrates identity without hegemony; the Assyrian and Hittite empires demonstrate hegemony over populations that did not regard themselves as one people with their rulers. It follows that evidence for hegemony is not evidence for identity, and the standard framing — “was there a Mycenaean empire?” — obscures the more interesting of the two questions.
A further complication attends claim 2. Power can be real and unnamed. Informal primacy among elites who exchange gifts, intermarry, and campaign together leaves no titulary trace. A finding that no superordinate title existed would therefore refute claim 2 without refuting claim 1. The strongest form of the minimalist position must accordingly rest not on the absence of titles but on the absence of transactions — of recorded flows of goods, personnel, or obligation running between kingdoms.
3. The Genre Problem
The most important methodological point is also the most often skipped: Linear B is unlikely to answer these questions, and its silence is therefore weaker evidence than it appears.
The extant corpus consists almost entirely of single-year palatial accounting — inventories, ration issues, personnel lists, landholding records, offering schedules, assessments of contributions in bronze, flax, wool, oil, and grain (Ventris & Chadwick, 1973; Killen, 2008; Nakassis, 2013). It is the residue of a fiscal year, preserved by the fires that ended it. There is no narrative text, no law, no letter, no treaty, no hymn, no genealogy, no king list. Whether such genres existed and were written on perishable media, or whether they existed only orally, or whether they did not exist at all, is undetermined.
The comparative evidence favors perishable media over non-existence. Waal (2011) has argued that Hittite Anatolia maintained a hieroglyphic scribal tradition on wooden writing boards alongside the cuneiform clay archives, leaving the wooden record almost entirely lost. A wooden diptych with ivory hinges was recovered from the Uluburun wreck, a vessel whose cargo touched the Aegean world (Pulak, 1998). If Aegean palaces wrote letters and agreements on wood, leather, or painted plaster, their absence from the clay record follows automatically and tells us nothing about their content.
Two consequences follow. First, the burden of the argument from silence must shift from “no such texts survive” to “no trace of such texts survives” — no references in the clay to sealed documents of another kind, no scribal vocabulary implying a second medium, no administrative gaps requiring one. Second, the highest-value hypothetical discoveries are not more accounting tablets but documents of new genre. A single Aegean letter would be worth a thousand ration lists.
4. What Would Have to Be Found: Hegemony
The following are graded by decisiveness. Tier 1 items would be individually close to conclusive; Tier 2 items would be strong but reversible; Tier 3 items would raise probability without settling anything.
Tier 1
(a) A superordinate title, multiply attested. A Linear B form standing above wa-na-ka — a Greek equivalent of “great king,” a construction of the type wanax wanaktōn, or a distinct lexeme applied to one ruler and never to others — appearing at two or more sites, in two or more scribal hands, in contexts where the grammatical relation to named subordinates is legible. The multiple-site and multiple-hand conditions are essential; a hapax legomenon in a partially understood syllabary is a lottery ticket, not a finding.
(b) A sworn agreement between kingdoms. The Late Bronze Age Near East produced a rich treaty literature: vassal treaties with historical preambles, stipulated obligations, troop levies, divine witness lists, and curse formulae (Liverani, 2001; Beckman, Bryce, & Cline, 2011). An Aegean document of this type — even a fragment preserving a witness list of gods and the names of two royal parties in asymmetrical relation — would settle claims 1 through 3 simultaneously. It is the single most valuable hypothetical text in the entire field.
(c) Correspondence with deference formulae. A letter, in any medium, from one Aegean ruler to another, using the address protocols that mark hierarchy in comparable corpora (“to my lord… thus says your servant”; cf. Moran, 1992, for the Amarna conventions). The presence of the letter genre at all would be a discovery; the presence of asymmetrical address within it would be decisive.
(d) Reciprocal obligation recorded at the subordinate end. A Pylos, Thebes, or Ayios Vasileios series recording annual assessed deliveries to Mycenae — using the allative in -de, as the tablets already do for internal destinations — booked as a recurring charge against districts rather than as a one-time transaction. Recorded from below, this establishes claim 3, which no assertion from the presumed center could establish alone.
Tier 2
(e) Shared personnel. The same named e-qe-ta or ko-re-te appearing in the archives of two kingdoms; or officials at one palace designated by an epithet indicating appointment from another; or a class of officials at multiple palaces bearing a title unattested as locally generated.
(f) Cross-polity military mobilization. The Pylian o-ka tablets record the disposition of coastal watchers under named commanders. A comparable set naming contingents drawn from other kingdoms, under a commander identifiable with a foreign center, would correspond structurally to the Catalogue of Ships and would give the epic tradition an administrative anchor.
(g) A named Aegean capital in an external archive. The Hittite Ahhiyawa texts never name the seat of the Ahhiyawan king. A text that did — pairing a Greek royal personal name with a recognizable toponym — would connect the diplomatic entity to a specific site and would constrain the debate sharply.
(h) Uniform royal sealing across kingdoms. Nodules or sealings at Pylos or Thebes struck from a device identifiable as the personal seal of a ruler seated elsewhere, in administrative rather than gift contexts.
Tier 3
(i) Convergent regnal dating across archives — documents at two palaces dated by reference to the same royal year. (j) A shared weights-and-measures authority recorded as externally imposed rather than merely shared in practice. (k) Records of judicial appeal or arbitration handled outside the kingdom of origin. The Pylian e-re-ta and land-dispute texts show internal legal process; external referral would imply a superior forum.
5. What Would Have to Be Found: Corporate Identity
The identity question requires different evidence, and much of it would be worthless for the hegemony question.
Tier 1
(a) A contrastive endonym. The decisive form is not a collective name in isolation but a collective name used against another. This point deserves emphasis, because the Pylos archive already supplies half of the pattern: groups of women are designated by foreign ethnics — Knidian, Milesian, and forms plausibly rendering Aswian and Lemnian origins — recorded as textile workers.² The palatial scribes possessed and used a category apparatus for foreigners. What is missing is the reciprocal term: a plural designation for their own people, deployed in a context where the contrast does the work. A ration list, a personnel roster, or a captive tally distinguishing “our people” from named outsiders would be nearly conclusive.
(b) A general term for outsiders. The Archaic and Classical barbaros is the mirror of Hellēn; each requires the other. A Bronze Age lexeme covering non-Greek-speakers as a class — rather than by specific origin — would establish that the in-group category existed, since a general out-group term cannot be formed without one.
(c) A multi-kingdom sanctuary record. Offering tablets from a cult site recording contributions labeled by kingdom of origin — Pylian, Theban, Argive deliveries to a single shrine, on a repeating schedule. This would be the Bronze Age equivalent of the Delphic amphictyony’s contribution structure and would establish corporate identity without requiring hegemony. It is the find that would most disturb the standard narrative, because it would place the mechanism that scholarship assigns to the Iron Age (Hall, 2002; Osborne, 2009) three centuries earlier.
Tier 2
(d) Aegean-authored diplomatic correspondence naming the sender’s land and people. A letter from an Aegean court to Egypt, Hatti, Ugarit, or Alashiya, in Akkadian, preserved in the recipient’s archive. Diplomatic exchange in this period used Akkadian as its medium; if the Aegean participated in the system at all, such letters were composed. This is the only genre in which a Bronze Age Greek ruler would have had occasion to say what his land and his people were called, in his own words, for an audience that needed to know. It bears on identity and hegemony at once.
(e) A cross-dynastic genealogy. Any text linking ruling houses of different kingdoms to a common ancestor — the function later performed by the Hesiodic scheme of Hellen and his sons (West, 1985; Hall, 1997). A Bronze Age king list with an ancestral apex shared between two centers would show the aggregative mechanism operating within the palatial period.
(f) Convergent festival or month names. The extant calendars of Knossos and Pylos differ. Attestation of a common festival, observed on a common date, across three or more archives would indicate a shared religious framework beyond the local pantheon.
Tier 3
(g) An itinerary or list of lands, showing how Aegean scribes classified the wider world and where they placed themselves within it. (h) Linguistic self-consciousness of any kind: a note on interpreters, a term for translated speech, a designation for intelligible versus unintelligible language. (i) A collective divine epithet — a deity styled as belonging to a people rather than to a place.
6. False Positives: What Would Look Decisive and Would Not Be
An evidentiary program is only as good as its list of traps.
Administrative uniformity. The similarity of formatting, ideograms, metrical systems, and scribal conventions across Knossos, Pylos, Thebes, and Mycenae is real and striking (Palaima, 2011). It is not evidence of political command. Scribal traditions propagate by training, and a common origin two centuries earlier accounts for the pattern without any continuing central authority. This datum is already in hand and is regularly over-read.
Prestige goods bearing Argolid iconography. Objects circulate by gift-exchange, dowry, trade, plunder, and craftsman mobility. Their distribution constrains contact networks, not sovereignty. The same caution applies in reverse to the wider “Mycenaean” material distribution, which grades outward from a core rather than terminating at a boundary (Feuer, 2011).
Synchronous destruction horizons. That many palaces fell within a few decades around 1200 BC is consistent with a shared systemic vulnerability — climatic stress, exchange-network failure, raiding, internal revolt, or all of these (Cline, 2014; Middleton, 2010). Simultaneity of collapse is not evidence of prior unity.
Homeric name-matching. Linear B already contains personal names corresponding to Hektor, Achilleus, Theseus, and Orestes, borne by ordinary men, some of them of unremarkable status. The onomastic overlap proves the antiquity of the name stock and nothing else. A tablet bearing a form readable as an antecedent of Agamemnon would be misreported within a day as confirmation of the Iliad, and it would confirm nothing except that the name existed. The methodological rule should be stated bluntly: names are evidence about names.
Exonyms treated as endonyms. Further Hittite or Egyptian texts naming Ahhiyawa or Tanaja more fully would refine the outside view. They would not tell us what the Aegean called itself. The distinction is the central finding of the companion paper and would survive a considerable enlargement of the external corpus.
The singleton. Any Tier 1 reading resting on one tablet, one hand, one site, and one plausible decipherment should be held provisionally regardless of how well it fits. The history of Aegean scholarship contains enough reversals to justify the caution.
7. Where Such Documents Could Survive
Specifying desiderata is empty without asking where they could physically be.
Unfound Aegean archives. No archive has been recovered from the citadel of Mycenae itself; the tablets from the site come largely from houses outside the walls and concern oil and textiles. Orchomenos, Iolkos/Dimini, Midea, and the Laconian and Messenian second-order centers remain incompletely explored. Ayios Vasileios is producing material now. The probability that additional palatial archives exist is high; the probability that they contain new genres is a separate and much lower question, since the same fire-preservation mechanism selects for the same kind of document.
Near Eastern archives. This is the more promising channel, precisely because it preserves the genres the Aegean lacks. Substantial portions of Hattusa remain unexcavated. Ugarit, Alalakh, Emar, Enkomi, Kition, and Delta sites such as Tell el-Dab’a are candidates for an Aegean-authored Akkadian letter. Any such find would arrive already legible, in a well-understood language, with its diplomatic conventions decodable against a large comparative corpus. One tablet of this kind would outweigh an entire new Aegean archive.
Wrecks. Written cargo manifests naming senders and recipients would establish who was transacting with whom at a state level. The Uluburun and Cape Gelidonya wrecks yielded no such document, but the wooden diptych from Uluburun indicates the medium existed in that world (Pulak, 1998).
Waterlogged and anoxic deposits. The only realistic route to Aegean texts on perishable media. Wells, cisterns, and harbor deposits are the plausible contexts. This is a low-probability, extremely high-value target class.
8. Four Counterfactual Scenarios
Scenario A: The Laconian Oath
Suppose: the Ayios Vasileios archive yields a set of tablets in a hand distinct from the local administrative scribes, recording sworn obligations of the Laconian wanax to a superior, with divine witnesses, a fixed annual delivery of bronze and men, and a clause on the return of fugitives.
What follows: claims 1, 2, and 3 are established for at least one relationship. The genre argument collapses: the Aegean did produce treaties on clay, and their absence elsewhere becomes an accident of preservation rather than a structural fact.
What does not follow: nothing about claim 4. A vassal relationship implies no shared identity — Ugarit’s subordination to Hatti did not make Ugaritians Hittites. Nor does one relationship generalize. It would remain possible that the Argolid held Laconia and nothing else.
Scenario B: The Aegean Letter
Suppose: a Levantine or Egyptian archive yields an Akkadian letter from an Aegean ruler, naming himself, his land, and his people, and addressing the recipient as brother.
What follows: the Aegean was a full participant in the great-king system; a self-designation is recovered; the relation of that self-designation to Achaioi, Danaoi, and Ahhiyawa becomes testable. Depending on the terms of address, the sender’s rank within the system is fixed.
What does not follow: that the name he gives covers all Aegean Greek speakers rather than his own realm. A ruler writing “the land of X” is naming his kingdom. Only a term applied to persons in the plural, contrastively, would establish claim 4 — and diplomatic letters rarely have occasion for it.
Scenario C: The Common Sanctuary
Suppose: a cult site outside any palace’s core territory yields offering records naming contributions from four kingdoms across multiple years, with no indication of compulsion.
What follows: claim 4 without claim 1. The Aegean possessed an institution of shared cult that transcended political boundaries, three centuries before Olympia. The standard model of Hellenic identity as a post-collapse achievement requires revision, and the sharp Bronze Age–Iron Age break in identity formation softens into continuity across the collapse.
What does not follow: political unity. This scenario would in fact strengthen the minimalist reading of hegemony while overturning the minimalist reading of identity — a combination that current framings barely contemplate, and the reason the two questions should be kept apart.
Scenario D: The Null Result
Suppose: three further palatial archives are recovered, each of the familiar type — internal, single-year, self-contained, with local officials, local deities, local calendars, and no reference to any external authority.
What follows: more than is usually granted. Each independent-looking archive raises the cost of the maximalist hypothesis, because a hegemony that generated no trace in four, five, or six separate administrations begins to require a mechanism of remarkable invisibility. The likelihood ratio matters more than any single find: what makes evidence powerful is how surprising it would be under the rival hypothesis. Archives that look exactly like Pylos are unsurprising under minimalism and increasingly awkward under maximalism.
This is worth stating plainly because null results are rarely published as findings. A field that specifies its expectations in advance can count them.
9. Standards of Proof
Three rules would improve the handling of whatever arrives.
Multiplicity. No Tier 1 claim on the basis of a single attestation. Two sites, two hands, and a context that constrains the reading independently of the desired conclusion.
Genre before content. Establish what kind of document is in hand before extracting propositions from it. A misidentified genre produces confident nonsense; an accounting notation read as a political assertion is the likeliest specific error.
Separation of the four claims. Every reported find should be assessed against each of the four claims separately and explicitly, including a statement of which claims it does not touch. Most over-reading in this field consists of evidence for claim 1 presented as evidence for claim 4, or the reverse.
10. Conclusion
The companion paper argued that Bronze Age Greece had no name for itself and probably no single ruler, and that the fragility of Agamemnon’s authority in the Iliad corresponds to that condition. Those conclusions rest on absences, and absences can be filled.
The point of specifying in advance what would fill them is not merely procedural hygiene. It is that the specification itself sharpens the original argument. Once one asks what a Bronze Age treaty, a Bronze Age contrastive ethnonym, or a Bronze Age common sanctuary ledger would look like, it becomes clear how much of the Near Eastern world’s ordinary documentary furniture the Aegean palaces appear not to have possessed — and clear also that the appearance may be an artifact of the medium that burned well. The honest position is that the minimalist reading is well supported by what survives and poorly protected against what might not have survived.
A discipline can hold that position with confidence only if it has said beforehand what would change its mind. This paper is an attempt to say so.
Notes
- Excavation at Ayios Vasileios (Xirokambi, Laconia) has proceeded under the Greek Archaeological Society since 2008 and has yielded Linear B material from a destruction context, with publication ongoing largely in Greek excavation reports. Any assessment offered here is provisional pending full publication of the archive.
- The Pylos textile-worker series includes group designations widely read as ethnics of Anatolian and island origin, including forms taken as Knidian and Milesian. Readings and identifications are debated; see Ventris and Chadwick (1973) and the discussion of workgroup terminology in Nakassis (2013). The methodological point stands even where individual identifications are uncertain: the scribes classified some persons by external origin, and did not classify anyone by internal collective origin.
- The condition that Tier 1 items be attested in multiple scribal hands is not pedantry. Palaima (2011) has shown how much of the Linear B corpus is the work of small numbers of identifiable individuals; a term used by one scribe may reflect that scribe’s practice rather than institutional vocabulary.
- The absence of the letter genre from the Linear B corpus is total. This is a genre absence of a different order from the absence of a particular word, and it is the strongest available argument that Aegean palatial writing was purely internal and administrative — or, alternatively, that external communication used a medium that has not survived. The two readings are not currently separable on the evidence.
- The “likelihood ratio” framing in §8 is used informally. No numerical treatment is intended; the relevant discipline is simply the habit of asking, of any proposed evidence, how expected it would be under the hypothesis one is inclined to reject.
- Scenario C is deliberately constructed to cross the usual battle lines, since a finding that supports minimalism on hegemony while overturning minimalism on identity would be difficult to assimilate to either standard position, and is for that reason a useful test of whether the field’s categories are doing analytical work or partisan work.
- On the wider methodological problem of reading archaeological and textual evidence against the Homeric tradition, Sherratt (1990) remains a useful corrective to confirmation-driven interpretation in both directions.
References
Aravantinos, V. L., Godart, L., & Sacconi, A. (2001). Thèbes: Fouilles de la Cadmée I. Les tablettes en linéaire B de la Odos Pelopidou — Édition et commentaire. Istituti Editoriali e Poligrafici Internazionali.
Beckman, G. M., Bryce, T. R., & Cline, E. H. (2011). The Ahhiyawa texts. Society of Biblical Literature.
Bennet, J. (2011). The geography of the Mycenaean kingdoms. In Y. Duhoux & A. Morpurgo Davies (Eds.), A companion to Linear B: Mycenaean Greek texts and their world (Vol. 2, pp. 137–168). Peeters.
Chadwick, J. (1976). The Mycenaean world. Cambridge University Press.
Cline, E. H. (2014). 1177 B.C.: The year civilization collapsed. Princeton University Press.
Cosmopoulos, M. B. (2019). State formation in Greece: Iklaina and the unification of Mycenaean Pylos. American Journal of Archaeology, 123(3), 349–380.
Davis, J. L., & Stocker, S. R. (2016). The lord of the gold rings: The Griffin Warrior of Pylos. Hesperia, 85(4), 627–655.
Dickinson, O. (2006). The Aegean from Bronze Age to Iron Age: Continuity and change between the twelfth and eighth centuries BC. Routledge.
Duhoux, Y., & Morpurgo Davies, A. (Eds.). (2008–2011). A companion to Linear B: Mycenaean Greek texts and their world (Vols. 1–2). Peeters.
Feuer, B. (2011). Being Mycenaean: A view from the periphery. American Journal of Archaeology, 115(4), 507–536.
Hall, J. M. (1997). Ethnic identity in Greek antiquity. Cambridge University Press.
Hall, J. M. (2002). Hellenicity: Between ethnicity and culture. University of Chicago Press.
Kelder, J. M. (2010). The kingdom of Mycenae: A great kingdom in the Late Bronze Age Aegean. CDL Press.
Kelder, J. M., & Waal, W. J. I. (Eds.). (2019). From LUGAL.GAL to Wanax: Kingship and political organisation in the Late Bronze Age Aegean. Sidestone Press.
Killen, J. T. (2008). Mycenaean economy. In Y. Duhoux & A. Morpurgo Davies (Eds.), A companion to Linear B: Mycenaean Greek texts and their world (Vol. 1, pp. 159–200). Peeters.
Liverani, M. (2001). International relations in the ancient Near East, 1600–1100 BC. Palgrave.
Middleton, G. D. (2010). The collapse of palatial society in LBA Greece and the postpalatial period. Archaeopress.
Moran, W. L. (Ed. & Trans.). (1992). The Amarna letters. Johns Hopkins University Press.
Nakassis, D. (2013). Individuals and society in Mycenaean Pylos. Brill.
Osborne, R. (2009). Greece in the making, 1200–479 BC (2nd ed.). Routledge.
Palaima, T. G. (2011). Scribes, scribal hands and palaeography. In Y. Duhoux & A. Morpurgo Davies (Eds.), A companion to Linear B: Mycenaean Greek texts and their world (Vol. 2, pp. 33–136). Peeters.
Pulak, C. (1998). The Uluburun shipwreck: An overview. International Journal of Nautical Archaeology, 27(3), 188–224.
Shelmerdine, C. W. (Ed.). (2008). The Cambridge companion to the Aegean Bronze Age. Cambridge University Press.
Sherratt, E. S. (1990). “Reading the texts”: Archaeology and the Homeric question. Antiquity, 64(245), 807–824.
Ventris, M., & Chadwick, J. (1973). Documents in Mycenaean Greek (2nd ed.). Cambridge University Press.
Waal, W. J. I. (2011). They wrote on wood: The case for a hieroglyphic scribal tradition on wooden writing boards in Hittite Anatolia. Anatolian Studies, 61, 21–34.
West, M. L. (1985). The Hesiodic Catalogue of Women: Its nature, structure, and origins. Clarendon Press.
Wiener, M. H. (2007). Homer and history: Old questions, new evidence. In S. P. Morris & R. Laffineur (Eds.), EPOS: Reconsidering Greek epic and Aegean scripts (Aegaeum 28, pp. 3–33). Peeters.
