The Journal of Hellenic Studies (2022), 142, 219–254

doi:10.1017/S0075426922000131

RESEARCH ARTICLE

Deconstructing the Phoenician myth: ‘Cadmus and the

palm-leaf tablets’ revisited

Willemijn Waal
Leiden University, The Netherlands Institute for the Near East
E-mail: [email protected]

Abstract
According to Herodotus, the expression phoinikeia grammata means ‘Phoenician letters’ and refers to
the Phoenician origins of the Greek alphabet. This account has found general acceptance, but it is not
the only interpretation possible and other theories circulated in antiquity. The adjective φοῖνιξ does
not only mean ‘Phoenician’; it can also refer to a palm tree or the colour red. This article argues that
the expression phoinikeia grammata did not originally refer to the alphabet, but to Linear B writing on
palm leaves, as already suggested by Frederick Ahl. It is shown that the account of Herodotus is the
result of a ‘learned reinterpretation’, triggered by the ambiguity of the word φοῖνιξ. The new under-
standing of phoinikeia grammata proposeds here has some important consequences: it implies that
Linear B, as has long been suspected, was primarily written on palm leaves (hence ‘palm-leaf writ-
ing’) and that in classical antiquity there was at least a limited historical awareness of the existence
of pre-alphabetic writing systems in the Aegean. This paper adduces additional evidence to substan-
tiate these claims. Lastly, a case will be made that Herodotus’ incorrect reinterpretation has led to
the ‘Phoenicianization’ of Cadmus, who was originally a Greek hero.

Keywords: phoinikeia grammata; introduction of the Greek alphabet; Linear B writing; Greek historical
awareness; Cadmus

There is nothing more deceptive than an obvious fact.

Sherlock Holmes1

The expression phoinikeia grammata is usually translated as ‘Phoenician letters’. This

interpretation goes back to Herodotus, who linked phoinikeia grammata to the
Phoenician origins of the Greek alphabet. Classical scholars and others have generally
accepted his account, which has become the mainstay of the widespread notion that
the Greeks took over the alphabet from the Phoenicians in the ninth or early eighth cen-
tury BC. There are compelling reasons, however, to question Herodotus’ explanation. His
account is internally inconsistent and presents chronological difficulties. It was also not
the only, nor the oldest, theory about the introduction of the alphabet that circulated in
antiquity. The enigmatic word φοῖνιξ can have a myriad of meanings in Greek; it does not
just mean ‘Phoenician’; it may, for instance, also refer to a palm tree or the colour red.
The first part of this article will take a fresh look at the meaning of the expression phoi-
nikeia grammata. It will examine the various existing traditions about the introduction of

1
Arthur Conan Doyle, The Boscombe Valley Mystery (1891).

© The Author(s), 2023. Published by Cambridge University Press on behalf of the Society for the Promotion of Hellenic Studies.
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (https://
creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution, and reproduction in any medium,
provided the original work is properly cited.

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 220 Willemijn Waal

writing and phoinikeia grammata (sections I–II), and its attestations in classical texts
(sections III–V). In section VI, it is concluded that phoinikeia grammata originally did
not refer to the alphabet, but to Linear B writing on palm leaves and that the account
of Herodotus is to be understood as the result of a ‘learned reinterpretation’. In the second
part of the article, the main consequences of this new insight will be addressed. First, addi-
tional evidence will be presented to corroborate the hypothesis that the Linear B script
was not written primarily on clay, but rather on perishable materials, notably palm leaves
(section VII). In sections VIII–XI it will be argued that in classical antiquity there was, at
least in some (learned) circles, historical awareness of the existence and use of pre-
alphabetic writing systems in the Late Bronze Age Aegean, including a discussion of
the terms σήματα and γράμματα and ‘Pelasgic letters’. Next, the possible motives for
Herodotus’ false reinterpretation (XII) and its implications for the origins of Cadmus,
the legendary bringer of phoinikeia grammata, will be discussed (XIII). Sections XIV–XV
offer a synopsis of the main arguments presented in this article and their implications.
Finally, the Appendix provides a succinct discussion of the epigraphic attestations that
have been linked to phoinikeia grammata.

I. Herodotus and phoinikeia grammata

Herodotus (5.58.1–2) gives the following account of the introduction of the alphabet to
Greece:

οἱ δὲ Φοίνικες οὗτοι οἱ σὺν Κάδμῳ ἀπικόμενοι, τῶν ἦσαν οἱ Γεφυραῖοι, ἄλλα τε πολλὰ
οἰκίσαντες ταύτην τὴν χώρην ἐσήγαγον διδασκάλια ἐς τοὺς Ἕλληνας καὶ δὴ καὶ
γράμματα, οὐκ ἐόντα πρὶν Ἕλλησι ὡς ἐμοὶ δοκέειν, πρῶτα μὲν τοῖσι καὶ ἅπαντες
χρέωνται Φοίνικες· μετὰ δὲ χρόνου προβαίνοντος ἅμα τῇ φωνῇ μετέβαλλον καὶ
τὸν ῥυθμὸν τῶν γραμμάτων. περιοίκεον δὲ σφέας τὰ πολλὰ τῶν χώρων τοῦτον τὸν
χρόνον Ἑλλήνων Ἴωνες· οἳ παραλαβόντες διδαχῇ παρὰ τῶν Φοινίκων τὰ
γράμματα, μεταρρυθμίσαντες σφέων ὀλίγα ἐχρέωντο, χρεώμενοι δὲ ἐφάτισαν,
ὥσπερ καὶ τὸ δίκαιον ἔφερε, ἐσαγαγόντων Φοινίκων ἐς τὴν Ἑλλάδα, Φοινικήια
κεκλῆσθαι.
These Phoenicians, who came with Cadmus, including the Gephyraeans, settled in this
land and transmitted to the Hellenes, among many other kinds of learning, the alpha-
bet, which I believe the Greeks did not have before, but which was originally used by all
Phoenicians. As time went on the sound and the form of the letters changed. At this
time, the Greeks who lived around them were for the most part Ionians; they were the
ones who were taught the letters by the Phoenicians, and after making a few changes to
their form, put them to use and called these characters ‘Phoenician’—which was only
just, since the Phoenicians had brought them into Hellas.

After a short exposé on the use of skins instead of papyri by the Ionians, Herodotus then
goes on to say that he has seen ‘Cadmeian letters’, which in his view greatly resemble
Ionian letters (5.59):

εἶδον δὲ καὶ αὐτὸς Καδμήια γράμματα ἐν τῷ ἱρῷ τοῦ Ἀπόλλωνος τοῦ Ἰσμηνίου ἐν
Θήβῃσι τῇσι βοιωτῶν, ἐπὶ τρίποσι τρισὶ ἐγκεκολαμμένα, τὰ πολλὰ ὅμοια ἐόντα
τοῖσι Ἰωνικοῖσι.
I myself have seen Cadmeian characters in the temple of Ismenian Apollo at Boeotian
Thebes, engraved on three tripods, and for the most part looking like Ionian letters.

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 The Journal of Hellenic Studies 221

Herodotus relates that one of these three tripods dates to the time of Laius, the great-
grandson of Cadmus, one to the time of Oedipus, son of Laius and one to the time of
Laodamus, son of Eteocles. Though the historicity of these mythical figures may now
be doubted, they were regarded as historical persons by Herodotus. He estimated that
Cadmus had lived before the Trojan War, some 1,000 years before his own time (2.145).
Though there are good reasons to believe that the Greek alphabet was introduced much
earlier than is currently assumed by most, such an early introduction is scarcely credible,
and the reference to ‘Phoenicians’ in (pre-)Trojan War times is anachronistic.2 It is also
hard to imagine that such very ancient letters would have looked so similar to the script
in Ionia of Herodotus’ time.
In addition to the chronological difficulties, the manner in which Herodotus frames the
story calls for caution; he is known to use eyewitness reports to introduce material he
knew was controversial.3 As may be derived from the remark ‘I believe’ (ὡς ἐμοὶ
δοκέειν), Herodotus was aware of the existence of contrasting views. A further reason
for wariness is the fact that Herodotus’ discussions of inscriptions are often problematic.
Archaeological discoveries in recent decades have revealed at least two examples of
inscriptions that were plainly misunderstood by Herodotus, the Karabel inscription in
western Anatolia (2.106) and the dedication of Croesus at Delphi (1.52).4 The ‘Cadmeian’
dedications described by Herodotus in book 5 may likewise have been misinterpreted.
Quite possibly, the inscriptions Herodotus claims to have seen were not ancient inscrip-
tions from a distant past, but rather dedications from a much later period, dating to some-
where between the eighth and sixth centuries BC.5
Though the above difficulties with the account of Herodotus and, more generally, the
problems inherent to the use of Herodotus’ work for historical purposes, are well-known,6
the gist of his exposition on phoinikeia grammata is generally accepted.7 Considering the
above, however, there are sufficient reasons to reassess the argumentation of
Herodotus, and to explore alternative explanations.

2
For a different view, see, for example, Bernal (1990), who assumes a very early introduction of the alphabet to
Greece in the 14th century BC.
3
Dewald (1987) 155–59; see also Hornblower (2013) 179.
4
Herodotus claims that the Karabel relief shows the Egyptian pharaoh Sesostris and that the accompanying
inscription is written in Egyptian hieroglyphs. These are, however, Anatolian hieroglyphs, and the decoding of the
text by David Hawkins (1998) has shown that Herodotus’ translation is completely fictitious. The inscription that
Herodotus ascribes to the legendary king Croesus was discovered in 2005; see Papazarkadas (2014). As convinc-
ingly argued by Peter Thonemann (2016), the dedication was not made by the Lydian king, but by a lesser-known
Athenian individual named Croesus, which was quite a common name, who had died in battle. This misinterpre-
tation subsequently formed the basis of Herodotus’ narrative about the testing of the oracle of Amphiaraus by
Croesus, which can now be dismissed as fiction; see Thonemann (2016) 165 and recently Van Rookhuizen (2019)
25–27.
5
Marinatos (1958) 226; S. West (1985) 292 with references; Symeonoglou (1985) 101–02; Thonemann (2016) 160,
164. Note that Symeonoglou (1985) 101 solves the chronological conundrum by assuming that the Cadmeian let-
ters refer to the old city Cadmea/Kadmeia and not to the person Cadmus.
6
See, for example, Carpenter (1935) 7–8; Ahl (1967) 190; Fehling (1971) 102–04; and especially S. West (1985)
289–95.
7
The fact that his explanation has never been seriously questioned should be seen in the light of the long
tradition in classical scholarship that perceived Herodotus as a reliable and trustworthy source; see recently
Van Rookhuijzen (2019) 30–33. An additional reason may be the fact that the Phoenician background of the
Greek alphabet appears to be confirmed by other evidence, such as the Greek letter names and letter shapes,
which are obviously Semitic, and the omnipresence of the Phoenicians in the archaeological records from the
ninth century BC onwards, see also below (section XII). For the alleged occurrence of φοῖνιξ in relation to writing
in the epigraphic material, which has been adduced as ‘proof’ of Herodotus’ account of the Phoenician background
of the alphabet, see, for example, Hornblower (2013) 178, and also the Appendix (Supplementary Material).

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 222 Willemijn Waal

II. Alternative traditions about the introduction of writing to Greece and

phoinikeia grammata
The account of Herodotus is not the first, nor the only, ancient narrative about the intro-
duction of writing to Greece and the meaning of the expression phoinikeia grammata.8 The
first known theory about the origins of writings is attributed to Stesichorus (seventh–sixth
century BC), according to whom the Greek hero Palamedes was the inventor of writing.
Lilian Jeffery has suggested that Stesichorus refers to an already existing tradition,9 which
is very probable, though not certain.10 Next are the accounts of Hecataeus and Herodotus,
which date to the fifth century. Hecataeus argues that Danaus brought the letters from
Egypt, a view that is also found in some later authors.11 The idea that it was Cadmus with
the Phoenicians who introduced writing to Greece is first found in Herodotus, but it may
have existed before him.12 This theory was quite popular and taken up by various later
authors. Most, however, also mention alternative theories and it appears that the ancients
were divided about the origins of the Greek alphabet.13 As mentioned above, the remark ‘I
believe’ (ὡς ἐμοὶ δοκέειν) suggests that Herodotus himself was also aware of the existence
of conflicting views, and Stephanie West has plausibly suggested that his account may have
to be interpreted as a contribution to a current controversy.14
In general, one can distinguish traditions that assert a foreign origin of writing
(Phoenician, Egyptian, Assyrian) with Danaus, Cadmus, various individuals called
Phoenix or the Phoenicians as inventors or mediators on the one hand15 and, on the other,
traditions that ascribe the discovery to a Greek god or hero. Named as inventors are
Prometheus, the Muses, Hermes, Sisyphus, Palamedes, Pythagoras and Simonides, and,
from at least the fifth century onwards, Orpheus, Linus and Musaeus. In this context, ref-
erence is sometimes made to ‘Pelasgic letters’.16 All inventions are situated in the Heroic
Age, which is usually associated with the Late Bronze Age.
It may be clear from this very succinct overview that though the theory of Herodotus
gained most acceptance, in antiquity it coexisted with other narratives. Of interest is the
observation of Jean Schneider that it is surprising that the first surviving account at our
disposal, namely that of Stesichorus, does not mention the Phoenician background of the
Greek alphabet at all.17 If one accepts the conventional model according to which the
Phoenicians introduced the alphabet to Greece in the early eighth century, this would have
been a relatively recent event and one might expect some reference to it. By contrast, all
accounts, without exception, place the invention of writing much earlier, in the distant
legendary past. With this in mind, let us now turn to the alternative explanations that
have been adduced to account for the expression phoinikeia grammata.

8
For an excellent and thorough discussion of the wide array of the various traditions, see Ceccarelli (2013)
63–89 with appendix 2, and also Schneider (2004) and Jeffery (1967).
9
Jeffery (1967) 152.
10
Ceccarelli (2013) 74.
11
See, for example, the accounts of Pythodorus, Phillis of Delos, Anaximander and Dionysius.
12
See, for example, Gomme (1913a) 62; Jeffery (1967); S. West (1985).
13
Cf. Nadaff 2003 (43). See, for example, Diodorus Siculus (first century BC), who mentions the Phoenicians and
Cadmus, but states that the Syrians were the inventors, and the Phoenicians only changed letters. Plutarch (first–
second century AD) mentions not only Cadmus, but also Palamedes and Simonides. The Latin authors Tacitus and
Pliny also relate several different accounts. For a useful overview of the main Greek sources on the invention of
writing, see Ceccarelli (2013) appendix 2.
14
S. West (1985) 290.
15
Ceccarelli (2013) 63–74.
16
For the expression ‘Pelasgic letters’, see below, section XI.
17
Schneider (2004) 124.

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 The Journal of Hellenic Studies 223

In ancient Greek, the chameleonic word φοῖνιξ is open to multiple interpretations,

ranging from mythical birds to diseases and musical instruments.18 Liddell and Scott list
no less than 12 different meanings with derivations.19 The most well-known uses of φοῖνιξ
that are relevant for the present investigation include:

• ‘Phoenician’, ‘Carthaginian’20
• ‘purple’ or ‘crimson’
• ‘date palm’, ‘palm’

These three meanings are already attested in Homer, where the word occurs as a personal
name (Φοῖνιξ) and is used to refer to palm trees and Phoenicians, as well as to the colour
red.21 The fact that the word φοῖνιξ has so many meanings has led to various different
interpretations of the expression phoinikeia grammata. Apart from the above-discussed
interpretation found in Herodotus, ancient scholars connected phoinikeia grammata to
the colour red and to the use of palm leaves as writing material:22

Eὐφρόνιος δέ, ὅτι μίλτῳ τὸ πρότερον ἐγράφοντο, ὅ ἐστι χρῶμά τι φοινικοῦν·

Ἐτεωνεὺςδὲ καὶ Mένανδρος, ἐπειδὴ ἐν πετάλοις φοινικείοις ἐγράφοντο· ἤ, ὅπερ
κρεῖττόν ἐστιν εἰπεῖν, ὅτι φοινίσσεται ὑπ’ αὐτῶν ὁ νοῦς ἤγουν λαμπρύνεται.23
Euphronius however says, because earlier they used to write with a red ochre called
miltos [red lead: Plin. HN 33.115], which has a reddish colour; and Eteoneus and
Menander [BNJ 783 F 5], because they used to write on palm leaves; or, a better expla-
nation, because the mind is reddened by them, that is, is brightened.24

18
With respect to its adjectival use, not only the meaning but also the spelling is hybrid: apart from φοῖνιξ used
as an adjective, forms like φοινίκεος, φοινίκειος, φοινικήιος and φοινικικός are attested. No systematic pattern
can be detected in the distribution of these various spellings in the attestations discussed here. The ancient Greeks
appear to have used these forms without distinction for the various possible meanings.
19
LSJ s.v. φοῖνιξ. The etymology of φοῖνιξ is fraught with uncertainty and widely diverging proposals have been
made about its origins. Some scholars interpret it as a Greek word, whereas others consider it an oriental loan-
word; for discussion, see, for example, Beekes (2010) 1584 s.v. φoινóς; Edwards (1979) 94; Van den Broek (1972) 61–
65; Chantraine (1968) 1217–20. From antiquity onwards, various, at times wild, theories have been proposed to
explain the internal relations between the homonymic bird, palm tree and colour; see Van den Broek (1972) 51–56.
It is not my aim to reopen this debate, as this article is only concerned with the use of the word φοῖνιξ in con-
nection to writing.
20
Note that the term ‘Phoenician’ is a Greek invention, it was not used by the Phoenicians themselves. For a
discussion of the term ‘Phoenician’ and the ‘Phoenician’ identity, see recently Quinn (2017) and the important
review of Gzella (2018), and for the term ‘Phoenician script’, see Lehmann (2019), who argues for the use of the
terms ‘Early Alphabetic C’ or ‘post Proto-Canaanite’ instead.
21
In Hesiod, it might refer for the first time to a mythical bird, though this remains doubtful, as the evidence is
quite ambiguous (Hes. fr. 304 Merkelbach-West). It is mentioned in the following riddle: ‘A screaming crow lives
for nine generations of men who have reached puberty; a deer is four crows; the raven grows old at three deer;
then the phoenix at nine ravens; and we at ten phoenixes, we beautiful-haired nymphs, daughters of aegis-
holding Zeus’ (fr. 254, tr. Most (2018) 361–63). Considering the fact that in the preceding lines animals are men-
tioned (a stag and two birds), it makes sense to imagine that ‘phoenix’ here refers to an animal as well. However,
as the legendary phoenix rising from its ashes has eternal life, its presence is rather unexpected in this context. It
should therefore not be excluded that ‘phoenix’ here in fact refers to a palm tree, which were known to live a very
long time.
22
More eccentric explanations include the idea that the word phoinikeia is related to phōnē, because the letters
represent the written voice, or that scribes used an instrument made of palm wood; see Jeffery (1967) 157–58 with
references.
23
Schol. Dion. Thrax 6. 20–25 Hilgard.
24
Tr. Ceccarelli (2013) 359.

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 224 Willemijn Waal

The explanation that the expression refers to ‘red letters’ because they ‘reddened’ the
mind fails to convince; could this be an addition of the scholiast himself, as suggested by
Arnold Gomme?25 The connection with red ink, however, has received some attention in
modern scholarship.26 Right between the two explanations connecting phoinikeia grammata
to the colour ‘red’, it is mentioned that according to the authors Etenoneus and Menander
the expression is related to the practice of writing on palm leaves. This explanation is also
found in the lexicon of Photius and the Suda, the well-known Byzantine encyclopaedic
work from the tenth century AD.27 The entry Φοινικήια γράμματα in the Suda reads as
follows:

Λυδοὶ καὶ Ἴωνες τὰ γράμματα ἀπὸ Φοίνικος τοῦ Ἀγήνορος τοῦ εὑρόντος· τούτοις δὲ
ἀντιλέγουσι Κρῆτες, ὡς εὑρέθη ἀπὸ τοῦ γράφειν ἐν φοινίκων πετάλοις. Σκάμων δ’ ἐν
τῇ δευτέρᾳ τῶν εὑρημάτων ἀπὸ Φοινίκης τῆς Ἀκταίωνος ὀνομασθῆναι. μυθεύεται δ’
οὗτος ἀρσένων μὲν παίδων ἄπαις, γενέσθαι δὲ αὐτῷ θυγατέρας Ἄγλαυρον, Ἔρσην,
Πάνδροσον· τὴν δὲ Φοινίκην ἔτι παρθένον οὖσαν τελευτῆσαι. διὸ καὶ Φοινικήϊα τὰ
γράμματα τὸν Ἀκταίωνα, βουλόμενόν τινος τιμῆς ἀπονεῖμαι τῇ θυγατρί.
Lydians and Ionians [call] the letters [thus] from their inventor Phoinix the son of
Agenor; but Cretans disagree with them, [saying that] the name was derived from
writing on palm leaves. But Skamon in his second book on Discoveries [says] that they
were named for Phoinike the daughter of Aktaion. Legend tells that this man had no
male children, but had daughters Aglauros, Erse, and Pandrosos; Phoinike, however,
died while still a virgin. For this reason, Aktaion [called] the letters Phoenician,
because he wanted to give some share of honour to his daughter.28

Likewise, in the scholia to Dionysius Thrax reference is made to writing on palm leaves
alongside other explanations:

Tινὲς δέ φασι τοὺς χαρακτῆρας τῶν στοιχείων τοὺς παρ’ ἡμῖν ὑπὸ Ἑρμοῦ ἐν φοίνικος
φύλλῳ γεγραμμένους καταπεμφθῆναι τοῖς ἀνθρώποις, διὸ καὶ φοινίκεια λέγεται τὰ
γράμματα· οἱ δέ, ὅτι Φοινίκων ἐστὶν εὕρεσις· οἱ δέ, ὅτι ὁ παιδαγωγὸς τοῦ
Ἀχιλλέως ὁ Φοῖνιξ ἐφεῦρεν αὐτά.29
Some say that the shapes of the elements which we use were transmitted to mankind
by Hermes, written on a palm-leaf, which is why the letters are called phoinikeia;
others however say they are a discovery of the Phoenicians; yet other, that the pae-
dagogue of Achilles, Phoenix invented them.30

In sum, several conflicting theories existed about the meaning of phoinikeia grammata.
The ‘Phoenician’ interpretation advocated by Herodotus may have become the most pop-
ular, but it did not preclude other narratives, according to which the expression originally
referred to ‘red letters’, or to writing on palm leaves. These alternative explanations stem
from a much later time and should therefore be treated with caution, as they could be the
result of later reinterpretations and/or inventions. By the same token, however, the fact

25
Gomme (1913a) 61.
26
See Mylonas (1966) 204 and the Appendix for further references. This explanation is also found in Isidore of
Seville (sixth–seventh century AD).
27
φ 652 Porson, φ 787 Adler.
28
Tr. Roth (2002). Note that the connection with Crete is also found in the scholia to Diodorus Siculus, where it
is stated that according to Dosiades the letters are an invention of Crete: Δοσιάδης δὲ ἐν Κρήτῃ φησὶν εὑρεθῆναι
αὐτά (τα γράμματα) (FGrH/BNJ 458 F6); see Ceccarelli (2013) 357–58.
29
Schol. Dion. Thrax 32. 9–13 Hilgard.
30
Tr. Ceccarelli (2013) 357.

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 The Journal of Hellenic Studies 225

that these theories are attested quite late, does not a priori make them false, nor does it
exclude the possibility that they are in fact much older; it is generally agreed that the Suda
and Photius preserve ancient knowledge, if often in garbled form. It is therefore worth-
while to reassess the primary attestations of phoinikeia grammata to see whether the alter-
native explanations may offer a more cogent interpretation than the commonly accepted
narrative.

III. Phoinikeia grammata used to mean ‘Phoenician letters’ or ‘red letters’

Most attestations of phoinikeia grammata are found in scholia.31 They mainly appear in dis-
cussions about the origin and meaning of this expression, some of which have been treated
above. When examining the ‘primary’ usage of this expression, that is the ways in which it
occurs in classical sources, it appears that the combination phoinikeia grammata was by no
means standard; the noun γράμματα (or, alternatively, στοιχεῖα) is usually used by itself,
without the adjective φοῖνιξ, to refer to ‘letters’.32 The same applies to the Latin expression
litterae Punicae/Phoenicae. The addition of the adjective provided extra, significant informa-
tion and its meaning depended on the context. Three main categories can be distinguished.
First of all, in some instances, the adjective φοῖνιξ (Latin: punicus) has to be understood as
an ethnikon, referring to ‘Phoenician/Punic’ letters. For example, Livy (28.46.16–18) relates
that Hannibal erected an altar recording his deeds in both the Greek and Phoenician
alphabets:33

propter Iunonis Laciniae templum aestatem Hannibal egit, ibique aram condidit ded-
icavitque cum ingenti rerum ab se gestarum titulo Punicis Graecisque litteris
insculpto.
Hannibal spent the summer near the temple of Juno Lacinia, and there he erected an
altar and dedicated it with a very long record engraved in Punic and Greek characters,
setting forth the achievements he had performed.34

Likewise, when Cicero (106–143 BC) talks about litterae punicae (Verr. 2.4.103) in connection
with the North African king Masinissa, he is referring to the contemporary alphabet used
there. Upon discovering that they were stolen from a temple, Masinissa sent back some
large teeth that were given to him, with an inscription:

itaque in iis scriptum litteris Punicis fuit regem Masinissam imprudentem accepisse,
re cognita reportandos reponendosque curasse.
and there was engraved on them in Punic characters that Masinissa the king had
accepted them imprudently; but, when he knew the truth, he had taken care that
they were replaced and restored.

An example from a much later period is provided by Theophanes the Chronographer

(ca. 752–818 AD), who describes two stelae in Libya inscribed with phoinikeia grammata by

31
For an overview and discussion of these attestations, see Ceccarelli (2013) 63–89 with appendix 2; Schneider
(2004); Jeffery (1967).
32
For the meaning of γράμματα, see below, section IX. For other possible uses of φοῖνιξ in epigraphic sources,
see the Appendix.
33
This inscription at Lacinium is also reported at Polybius 3.33.18, where an inscription on a bronze tablet
containing the deeds of Hannibal is mentioned.
34
Translation based on Jal (1995) 104.

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 226 Willemijn Waal

fugitives, referring to the contemporary Phoenician or Punic alphabet in use there at

the time:

στήσαντες δύο στήλας ἐπὶ τῆς μεγάλης κρήνης ἐκ λίθων λευκῶν ἐγκεκολαμμένα
ἐχούσας γράμματα Φοινικικὰ λέγοντα τάδε· ἡμεῖς ἐσμέν οἱ φυγόντες ἀπὸ
προσώπου Ἰησοῦ τοῦ λῃστοῦ, υἱοῦ Ναυῆ.35
they erected two stelae at the large fountain made of white stone, containing carved
Phoenician characters which read as follows: we are the fugitives from the face of the
pirate Jesus, son of Nave.

Secondly, phoinikeia grammata may refer to ‘letters’ painted in red, as in the below pas-
sage from Cassius Dio (second–third century AD):

σημεῖον δέ τι τῶν μεγάλων, τῶν τοῖς ἱστίοις ἐοικότων καὶ φοινικᾶ γράμματα ἐπ’
αὐτοῖς πρὸς δήλωσιν τοῦ τε στρατοῦ καὶ τοῦ στρατηγοῦ σφων τοῦ αὐτοκράτορος
ἐχόντων, ἐς τὸν ποταμὸν ἀπὸ τῆς γεφύρας περιτραπὲν ἐνέπεσε.36
But one of the large flags that resemble sails with red letters upon them to distinguish
the army and its commander-in-chief, was overturned and fell from the bridge into
the river.37

The third and, for the present investigation, most interesting category is made up of
cases in which phoinikeia grammata refers to (very) ancient inscriptions.

IV. Phoinikeia grammata referring to ancient inscriptions

A well-known example of phoinikeia grammata referring to an ancient inscription is the
lebēs inscribed by Cadmus, which features in the Lindian Chronicle. This fascinating docu-
ment, which is written on a marble slab and dates to 99 BC, records the dedications made
to the temple of Athena at Lindos before the destruction of the original temple in 392/1 BC.
Among the votive objects from ancient times mention is made of a bronze lebēs of Cadmus,
which was inscribed in phoinikika grammata, according to Polyzalus:38

Κάδμος λέβητα χά[λ]κεον φοινικικοῖς γράμμα-

σι ἐπιγεγραμμένον ὡς ίστορεῖ Πολύζα-
λος ἐν τᾶι Δ ταν ἱστοριᾶν.39
Cadmus, a bronze lebēs. Inscribed with
phoinikika grammata as Polyzalus reports
in the fourth book of his Investigations.40

The historian Diodorus Siculus (first century BC) also mentions this inscribed bronze
lebēs of Cadmus in his description of Rhodes. Like the Lindian Chronicle, which may have

35
Theophanes, Chronographia I, de Boor 21–24.
36
Cass. Dio 40.18.3.
37
Translation based on Lachenaud and Coudry (2011) 162.
38
Not much is known about this author, who may have lived in the end of the fourth or beginning of the third
century BC; see Higbie (2003) 72 with references.
39
Lindian Chronicle B 15–17.
40
Translation based on Higbie (2003) 23.

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 The Journal of Hellenic Studies 227

been his source,41 Diodorus explicitly mentions that the inscription is written in phoinikeia
grammata (5.58.2):42

ὁ δ’ οὖν Κάδμος καὶ τὴν Λινδίαν Ἀθηνᾶν ἐτίμησεν ἀναθήμασιν, ἐν οἷς ἦν χαλκοῦς
λέβης ἀξιόλογος κατεσκευασμένος εἰς τὸν ἀρχαῖον ῥυθμόν· οὗτος δ’ εἶχεν
ἐπιγραφὴν Φοινικικοῖς γράμμασιν, ἅ φασὶ πρῶτον ἐκ Φοινίκης εἰς τὴν Ἑλλάδα
κομισθῆναι.
Cadmus also honoured Lindian Athena with votive offerings, among which there was
a bronze lebēs worthy of note, made in the archaic fashion. This had an inscription in
phoinikeia grammata, which they say were brought first from Phoenicia into Greece.43

Diodorus seems to take the expression phoinikeia grammata to refer to the alphabet, adding
that the letters were apparently first introduced to Greece from Phoenicia. From the inter-
jection ‘they say’, one can tell that for Diodorus this was not an established and undisputed
fact, and he was probably aware of other explanations.44
The assumption that phoinikeia grammata here refers to the alphabet, however, is not
self-evident. The Lindian Chronicle lists several inscribed objects. Only in two cases is the
type of script specified: in the above-quoted example and in the case of an inscription in
Egyptian hieratic (ἱερὰ γράμματα) by the Egyptian donor Amasis.45 The fact that the script
is further defined implies that these two inscriptions are different from the others (which
were presumably written in the Greek alphabet). Interestingly, they are also the only two
inscriptions in the chronicle of which the content is not quoted. This could either mean
that the compilers of the list could not read the inscriptions quoted by their source
(Polyzalus in the case of the dedication of Cadmus), or that their content was not quoted
by their source.46 One could come up with several explanations for this, but an obvious one
would be that the inscription could not be understood, because it was written in a different
script. This explanation is appropriate for the Egyptian dedication, which was after all
composed in foreign (hieratic) writing. What about the inscribed lebēs of Cadmus?47
One could argue that the inscription was written in the Phoenician alphabet, which the
compilers could not read.48 This, however, leaves us with the problem that an alphabetic
inscription was anachronistically associated with a hero from the Heroic Age.49 The
‘Phoenician’ interpretation is even more challenging in the following example.
The expression punicae litterae is mentioned in the Latin version of the so-called Diary or
Journal of Dictys of Crete. This is a document supposedly written by Dictys, the companion
of Idomeneus in the Cretan contingent at Troy, containing an eyewitness report of the

41
Higbie (2003) 72.
42
Note that a potentially similar attestation is found in an inventory from the Acropolis at Athens. IG II2 1456
(dated post-341/340 BC) mentions an ivory object inscribed with γράμματα φοινικικὰ (42–43: [γράμματ]α
φοινικικὰ ἔχο[ν]; compare the inventory IG II2 1485 (late fourth century BC), 38–39: [φοινικικ]ὰ γράμματ[α
ἔχον]). Unfortunately, the context is too uncertain to decide whether we are dealing here with an ancient inscrip-
tion, an inscription in the contemporary Phoenician alphabet or an inscription painted in red.
43
Tr. Higbie (2003) 72.
44
Cf. above, section I.
45
Lindian Chronicle C 51–53; Higbie (2003) 34–35.
46
Higbie (2003) 71.
47
It is improbable that phoinikeia grammata here refers to letters that were painted in red, as this is unlikely for
an inscription in bronze. Moreover, this would not clarify why the text is not cited.
48
Thus, for example, Willi (2005) 170, with reference to Guarducci (1987) 16.
49
For more on Cadmus and his alleged Phoenician origins, see below, section XII.

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 228 Willemijn Waal

Trojan War.50 It circulated in a Greek version in the late second century AD, of which some
papyrus fragments have survived.51 The Latin version, ascribed to the author Septimius
(third–fourth century AD), is accompanied by a prologue (prologus) as well as a prefatory
letter (epistula). The prologue is older than the letter and was probably translated from
Greek.52 In later editions, Septimius appears to have replaced this prologue by the prefa-
tory letter.53
The prologue opens by stating that Dictys, a Cretan by birth from the city of Knossos,
was a contemporary of the sons of Atreus and an expert in the ‘Phoenician’ language and
letters (peritus uocis ac litterarum Phoenicum). He wrote annals of the Trojan War in nine
volumes on linden-wood tablets in phoeniceis litteris, which he ordered to be buried with
him. After an earthquake in the 13th year of Nero’s reign, his tomb was laid bare and dis-
covered by shepherds. They saw the linden-wood tablets, which were covered with writing
that they did not recognize, and were subsequently brought to Nero:

haec igitur Nero cum accipisset aduertissetque punicas esse litteras harum peritos ad
se euocauit, qui cum uenissent, interpretati sunt omnia. cumque Nero cognosset anti-
qui uiri, qui apud Ilium fuerat, haec esse monumenta, iussit in Graecum sermonem
ista transferri e quibus Troiani belli uerior textus cunctis innotuit.
When, then, Nero had received them and recognized that they were in Phoenician
script, he called in experts in this script, who arrived and explained everything.
And when Nero realised this documented a man of long ago who had been at
Troy, he gave instructions for it to be translated into the Greek language, as a result
of which a truer account of the Trojan War became known to everyone.54

The later letter (epistula) of Septimius, which possibly accompanied the second edition
of the Latin version, gives a slightly different account:

pastores cum eo deuenissent, forte inter ceteram ruinam loculum stagno affabre clau-
sum offendere ac thesaurum rati mox dissoluunt non aurum nec aliud quicquam
praedae, se libros ex philyra in lucem prodituri. at ubi spes frustrata est, ad
Praxim dominum loci eos deferent, qui commutatos litteris Atticis—nam oratio
Graeca fuerat—Neroni Romano Caesari obtulit, pro quo plurimis ab eo donatus est.
Shepherds who arrived there [at the grave], by chance came upon a tin box among the
other rubble. So, thinking it was treasure they presently opened it. But what came to
light was not gold or anything profitable, but books of linden bark. As their expect-
ations had been disappointed, they took them to Praxis, the master of the place, who
transcribed them into Attic script—because it was in Greek—and took it to Nero, the
Roman Emperor, in return for which he got many gifts.55

Here, the documents are described as being written in Greek, but in a different script.
Earlier in the letter, Septimius states that the text was written in litteris Punicis, which were
introduced by Cadmus and Agenor and were quite widespread in the days of the
Trojan War:

50
For discussions of this intriguing text and its transmission, see Merkle (1989); Horsfall (2008–2009); Gainsford
(2012); Ní Mheallaigh (2012) 184; Dowden (2016).
51
Horsfall (2008–2009) 43 n.10.
52
Merkle (1989) 133; Horsfall (2008–2009) 44.
53
Horsfall (2008–2009) 44.
54
BNJ 49 T4.4, tr. Dowden (2016).
55
BNJ 49 T5.2, tr. Dowden (2016).

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 The Journal of Hellenic Studies 229

Ephemeridem belli Troiani Dictys Cretensis, qui in ea militia cum Idomeneo meruit,
primo conscripsit litteris Punicis, quae tum Cadmo et Agenore auctoribus per
Graeciam frequentabantur.
The Diary of the Trojan War was first written down by Dictys of Crete, who served in
that campaign with Idomeneus, in Phoenician script, which in those days, thanks to
Cadmus and Agenor, was widespread in Greece.56

From the scantily preserved earlier Greek version a similar picture emerges: it is men-
tioned that the text was ‘transcribed’ (μεταγραφῆναι, FGrH 49 T2c) after its discovery.57
The papyrus fragment P.Oxy. 4944 appears to relate that the text was written in what were
thought to be the letters of Cadmus and Danaus.58
Though the diary enjoyed the status of an authentic and authoritative narrative
(together with the diary of Dares) in the Middle Ages, modern scholarship views the
existence and rediscovery of this ancient document as a fabrication. The adage that
the more complex and specific the details of a text and its survival, the more they proclaim
its falsity, probably holds true,59 all the more because the Fundbericht shows all the formal
features of ‘pseudo-documentarism’.60 Since the ‘diary’ itself is held to be invented,
some scholars dismiss the punicae litterae as fictitious, arguing that they do not refer to
a historical, but rather a legendary script.61 Though a healthy dose of distrust is certainly
warranted, one should not too easily discard the account as mere fiction without any
historical relevance. The primary aim of the Beglaubigungsapparat built around this text
was to lend credibility to the fact that it was a truly ancient document, dating to the time
before Homer. In order to achieve the desired ‘reality effect’, the physical details, the
language and the script of the document (and its uncovering) had to sound convincing
to the audience.62 It has long been suggested that the report was inspired by genuine
discoveries of ancient documents.63
As for litterae punicae, they must have evoked an image in the minds of the audience,
which was plausible in this particular context. Karen Ní Mheallaigh attempts to answer the
question of what precisely the audience would have envisioned when hearing or reading
about a document composed in litterae punicae. She assumes that the document that is
described in the Latin prologue is not only written in Phoenician letters, but also in
the Phoenician language, though the latter is never explicitly stated anywhere. She argues
that readers were invited to think of Phoenician as the lingua franca in the heroic past. As
for the fact that Septimius’ letter mentions that the text was in Greek (nam oratio Graeca

56
BNJ 49 T5.1, tr. Dowden (2016).
57
Ní Mheallaigh (2012) 184.
58
P.Oxy. 4944: ταῦτα δ̣ὲ ἐγ̣[ὼ συνεγραψάμην,] | Δίκτυ[ς] Κνώσσι[ος . . . Κάδμου?] | καὶ Δα[ν]αοῦ γρά ̣[μμασιν. οὐ]
| γὰρ μιᾷ̣ χρῶντ[αι γλώσσῃ οὔτε] | πάντε̣ς οἱ Ἕλλη ̣ [νες οὔτε πάν]|τες οἱ βάρβαροι, ἀ[λλὰ
̣ μεμι]|γμένῃ. τοῦτο δ[ὲ
θαυμαστὸν] | μηδεὶς ἡγεῖσθ[ω εἶναι, ἐπεὶ] | καὶ ἡμεῖς οἱ ἐν [Κρήτῃ οὐ πάν|τ̣ες χρώμεθα τῇ ̣ [αὐτῇ γλώσσῃ], ‘I,
Dikty[s] of Knoss[os, wrote] this (history) . . . in the let[ters of Cadmus?] and Da[n]aus. For [neither] do all the
Gree[ks nor al]l the barbarians use a single [language], b[ut it is mix]ed. And no one should find this [surprising,
since] we too in [Crete] do [not a]ll use the [same language]’ (translation based on Gainsford (2012) 62 n.19). Since
this interpretation relies for the most part on restoration, it is best left out of the discussion. Note that, consid-
ering the context, it would make more sense that the heavily restored last two lines refer not to the existence of
various languages, but rather to the existence of various scripts, perhaps expressed by the noun γραφή.
59
Ní Mheallaigh (2008) 407–08; Horsfall (2008–2009) 45.
60
Hansen 2003; see also Ní Mheallaigh (2008) 406–14.
61
See, for example, Gainsford (2012) 63. Compare also Horsfall (2008–2009) 49, who considers the use of mys-
terious letters typical of the genre.
62
For the construction of the elaborate Beglaubigungsapparat for this text, see Hansen (2003) 304–08; Stott (2008)
93; Horsfall (2008–2009) 45–46; Gainsford (2012) 61–62; Higbie (2014) 13; Higbie (2017) 224–26.
63
See also below, section VIII.

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 230 Willemijn Waal

fuerat), and thus not in Phoenician(!), she speculates that Septimius here replaced the
‘Phoenician original’ with a ‘culturally more palatable Greek version’ to accommodate
the taste of the Roman audience. She admits that this solution is not very satisfactory,
and wonders why, if this were the case, Septimius did not ‘expunge the Journal’s distasteful
Punic pedigree entirely? It seems puzzling that Septimius should jettison the Phoenician
language Ur-text, yet retain the fiction of its Phoenician writing’.64 She solves this conun-
drum by suggesting that the ‘Punic letters’ represent a primitive form of the Greek alpha-
bet, linking it to its gradual development, which is a slightly forced and not entirely
consistent explanation.65
The idea that litterae punicae must refer to an ancient form of writing is appealing, how-
ever, especially if one recalls the inscription ascribed to Cadmus discussed above. Phoinikeia
grammata apparently represent a truly ancient script that was distinct from contemporary
writing, that was believed to have been in use before and during the Trojan War and, if we
attach any value to the remark in Septimius’ letter, that was used for the Greek language.
These facts combined make it very hard, if not impossible, not to think of earlier proposals
which, largely for entirely different reasons, link phoinikeia grammata to the Linear B
script.66 The ways in which the supposedly ancient Journal of Dictys is described (the fact
that it dates to the time of the Trojan War; the fact that the herdsmen did not recognize its
writing; the fact that it had to be transliterated and that it is in one account referred to as
Greek, whereas in the other it is mentioned that it had to be ‘transcribed’ or ‘translated’
into (Classical) Greek) fit all the characteristics of the syllabic Linear B script, which was
used for Mycenaean, an ancient dialect of Greek of the Late Bronze Age. Rather than a
‘puzzlingly, implausible Punico-Greek text’67 from a fanciful fantasy world dominated
by the Phoenician script and language, the alleged Journal of Dictys would instead
represent a document written in an attested ancient script from a real historic past, which
the audience rightly associated with the time of the Trojan War and before. As will be
argued further in detail below (section VIII), there is ample evidence that the Greeks were
aware of the existence of Late Bronze Age writing systems, either through archaeological
discoveries and/or (orally) transmitted stories. The assumption that phoinikeia grammata
refers to Linear B gains further strength if we look at other attestations of this expression,
which are also clearly situated in the Heroic Age, such as the story of Palamedes.
Palamedes was a Greek hero, (probably) identified as Argive, who according to some
traditions invented writing. He does not occur in the Homeric epics, but his legend is
undoubtedly ancient. Palamedes plays an important role in the Cypria, where he is the
one who uncovers the feigned madness of Odysseus. He also appears in the plays of
Aeschylus, Sophocles, Euripides and Astydamas, and the works of various other authors.68
In a scholion to Euripides’ Orestes, Palamedes is mentioned in connection with phoinikeia
grammata. The scholiast asserts that Palamedes solved the Achaeans’ food rationing prob-
lem during the Aegean fleet’s stay at Aulis on their way to Troy by showing them phoinikeia
grammata:

64
Ní Mheallaigh (2012) 187.
65
A similar suggestion has been made by Corcella (1986) with respect to the Pelasgic letters, see below,
section XI.
66
See, for example, Carpenter (1935) and Mylonas (1966) 204, who sees the connection in the fact that the
letters were painted in red, and Ahl (1967), who links phoinikeia grammata to palm-leaf writing (see also below,
section VI).
67
Ní Mheallaigh (2012) 182.
68
See Ceccarelli (2013) 76 with references.

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 The Journal of Hellenic Studies 231

πρῶτον μὲν τὰ Φοινίκια διδάξας γράμματα αὐτοὺς ἴσην τε καὶ

ἀνεπίληπτον τὴν διανομὴν ἐν τούτοις ἐπραγματεύσατο.69
First having taught them phoinikeia grammata, he concerned himself with the equal
and correct allocations among them.

It need not be pointed out that the distribution of food is precisely the type of usage
that is attested in the Linear B documents from the Late Bronze Age.70 Unfortunately, we
cannot be sure that the term phoinikeia was present in the original source, as it cannot be
entirely excluded that these are the words of the scholiast. In any case, as in the above
examples, phoinikeia grammata are again associated with the distant past.
Phoinikeia grammata are further mentioned in connection with the Trojan War in the
Poimenes of Sophocles, which we know only from a fleeting reference in Hesychius.71
Here, they seem to be mentioned in the context of the arrival of the Greeks at Troy.
The occurrence of ‘Phoenician letters’ here has puzzled some scholars,72 but it would
be much less surprising in light of the interpretation suggested here, in which phoinikeia
grammata refer to Linear B writing. We may further mention a fragment from Euripides’
Palamedes, in which the protagonist appears to defend his invention of writing:

τὰ τῆς γε λήθης φάρμακ҆ ὀρθώσας μόνος,

ἄφωνα φωνήεντα συλλαβὰς τιθεὶς,
ἐξηῦρον ἀνθρώποισι γράμματ҆ εἰδέναι.73
On my own I established remedies for forgetfulness, which are without speech and
(yet) speak, by creating syllables, I invented for mankind knowledge of writing.74

The description of Palamedes’ invention as ‘creating syllables’ makes perfect sense if this is
a reference to the Linear B script, which is of a predominantly syllabic nature, but is less
evident if one takes it to refer to alphabetic writing. However, since the precise translation
and reading of this passage is somewhat debated, this example should be treated with cau-
tion.75 Last but not least, the interpretation of phoinikeia grammata as Linear B writing
would elucidate a thus far opaque quotation from Timon cited by Sextus, which refers
to phoinikika sēmata (‘Phoenician signs’).

V. Sēmata phoinikika
In the book Against Grammarians, Sextus Empiricus (second–third century AD) recounts the
benefits of learning how to write and he counts literacy among one of the most useful
things. He then remarks the following:

69
Scholion on Orestes 432; see Schwartz (1887) 148 and also Ceccarelli (2013) 77.
70
Cf. Phillips (1957) 273.
71
Fr. 514 = φ 688.
72
Ceccarelli (2013) 205–06.
73
Kannicht (2004) 598, fr. 578.
74
Translation for the most part follows Collard and Cropp (2008b) 53.
75
Instead of φωνήεντα, the reading καὶ φωνοῦντα/φωνῆντα has been proposed: see Jouan and Van Looy (2000):
510; Kannicht (2004) 598 with references. Jouan and Van Looy (2000) 509–10 translate ‘en établissant les con-
sonnes, les voyelles et les syllabes, j’ai initié les hommes à la connaissance des lettres’. Collard and Cropp
(2008b) 53 n.2 give two options, proposing ‘consonants and vowels, by creating syllables’ and ‘by creating con-
sonants and vowels and syllables’. Scodel (1980) 91 n.27 takes ἄφωνα φωνήεντα here to have their later well-
attested meaning of ‘surds and sonants’, and Jenkins (2006) 20 translates ‘by organizing consonants and vowels
into syllables’.

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 232 Willemijn Waal

καίτοι δόξειεν ἄν τισιν ἐπὶ τῆς ἐναντίας εἶναι προλήψεως

ὁ προφήτης τῶν Πύρρωνος λόγων Tίμων ἐν οἷς φησι
γραμματική, τῆς οὔ τις ἀνασκοπὴ οὐδ’ ἀνάθρησις
ἀνδρὶ διδασκομένῳ Φοινικικὰ σήματα Κάδμου.76
However, some might think that Timon, the spokesman for Pyrrho’s discourses, has
the exact opposite preconception when he says: ‘Grammar, of which there is no con-
sideration nor inspection by a man being taught the Phoenician signs of Cadmus’.77

The quotation is attributed to Timon, who is known as the expounder of the work of the
‘sceptic’ Pyrrho of Elis (fourth–third century BC). Pyrrho does not appear to have written
down anything himself, but his works were recorded by his pupil Timon. Most of them
have been lost, and they are mainly preserved through the works of Sextus Empiricus.
The above quotation is not otherwise recorded and we thus have no idea of its original
context.78 Sextus is clearly struggling with the interpretation of this seemingly contradic-
tory saying of Timon. His deliberations are worth quoting in full:

But in fact, it doesn’t seem to be this way, for what he says, ‘there is no consideration
nor inspection’, is not such as to go against literacy itself, by way of which ‘the
Phoenician signs of Cadmus’ are taught, for if someone is being taught it, how has
he not made it his business? Rather, he is saying something like this: ‘for the person
who has been taught the Phoenician signs of Cadmus there is no business with any
other grammar beyond this’, which tends not towards this grammar—the one that is
observed in the elements and in writing and reading by means of them—being use-
less, but the boastful and busybody kind.
For the use of the elements bears directly on the conduct of life, but not to be satisfied
with what is handed down from the observation of these, and to demonstrate in addi-
tion that some are by nature vowels and others consonants, and that of the vowels
some are by nature short, others long and others two-timed, having length and short-
ness in common, and in general the rest of the stuff that the nonsense-filled gram-
marians teach—that is useless.79

According to Sextus, Timon here does not agitate against literacy itself, but rather
against boastful teachers.80 As kindly pointed out to me by one of the anonymous
reviewers, the explanation of Sextus should be understood in connection with the two
types of grammar that are mentioned earlier in the text:

However, since grammar is of two kinds, one professing to teach the elements
(στοιχεῖοv) and their combinations and being something of a general expertise in
reading and writing, the other being a deeper power than this, lying not in the bare
knowledge of letters (γράμμα) but also in the examination of their discovery and their
nature, as well as the parts of the discourse constructed from these and any other

76
Sext. Emp. Πρὸς γραμματικούς 53.
77
Tr. Bett (2018) 52–53.
78
Bett (2018) 53 n.48.
79
Sext. Emp. Πρὸς γραμματικούς 54, tr. Bett (2018) 53.
80
Note that Decleva Caizzi follows this interpretation and argues that the quotation should not be seen as a
rejection of literacy or poetry, as it is clear from other contexts that Timon does not object to these as such. She
suggests that Timon means to convey that it is possible to enjoy literature without the mediation of learned
grammarians (Decleva Caizzi (1981) 207, text 45).

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 The Journal of Hellenic Studies 233

object of the contemplation of the same sort, it is not our task to argue against the
first.81

Timon’s remark should be seen as referring to the second kind of grammar. This is, how-
ever, not evident from the quotation itself, which refers to grammar in general. Another
puzzling element of the quotation is that the word σήματα is used rather than expected
γράμματα.
If one takes phoinikika sēmata to refer to Linear B writing, an elegant solution to these
difficulties presents itself. First of all, it would explain the choice of the word ‘signs’
(σήματα) rather than ‘letters’ (γράμματα), which is a more adequate term to describe a
logosyllabic than an alphabetic writing system.82 More importantly, in this interpretation,
the quotation suddenly becomes comprehensible and any apparent contradiction or
ambiguousness dissolves. Timon’s argument is that literacy does not necessarily require
extensive knowledge about vowel length, etc., as one could in the past manage with a
logosyllabic writing system, which was unable to represent the relatively large number
of phonemic distinctions of Greek. Sextus is right to conclude that Timon is not attacking
literacy itself, but that he is militating against the (in his eyes) unnecessary phonetic rules
taught by grammarians. To illustrate his point, he contrasts the alphabet to the (in this
respect) much more defective logosyllabic Linear B writing system.
I am not implying that Timon or Pyrrho could actually read and understand Linear B,
but I do hold it to be conceivable that they were aware of its existence, either through later
discoveries of Linear B texts or through legends.83 Similarly, the ancient Greeks were
undoubtedly familiar with the concept of syllabic writing systems. Though the alphabet
may have been the dominant script in Classical Greece, it was not the only writing system
that was in use. In Cyprus, with which there were extensive contacts, the Cypriot syllabary,
ultimately a descendant of Linear A, existed till the fourth century BC. Inscriptions in the
Cypriot syllabary have turned up outside of Cyprus in various locations, including Lefkandi
(Euboea), the northern Aegean (Chalcidice), southern Italy (Policoro, Broglio di
Trebbesace), Sardinia and Delphi.84 From the pictorial appearance of Linear B texts,
one could easily derive that it had more in common with a syllabic than an alphabetic
script (as implied by the choice of the word σήματα), and that it was a writing system
with no, or little, regard for the kind of ‘stuff that the nonsense-filled grammarians teach’.
It is regrettable that we do not possess the original context of the quotation, but based
on the available evidence, the above interpretation clarifies this otherwise enigmatic and
seemingly self-contradictory phrase. At the same time, it eloquently demonstrates that to
Sextus the original meaning of phoinikeia sēmata/grammata was completely lost, and that
this knowledge certainly was not omnipresent in his time.

VI. Preliminary conclusions

The above overview has shown that the assumption that the expression phoinikeia gram-
mata refers to the Phoenician origins of the Greek alphabet as suggested by Herodotus
cannot be upheld. In some cases, its meaning is clear; it refers to the contemporary
Phoenician or Punic alphabet or to letters painted in red. In the other examples discussed,
the term phoinikeia grammata are is used in connection with ancient inscriptions and one
may summarize their characteristics as follows:

81
Sext. Emp. Πρὸς γραμματικούς 49, tr. Bett (2018) 52.
82
See also below, section IX.
83
See also below, sections VIII and X.
84
Bourogiannis (2019) 173–75 with references.

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 234 Willemijn Waal

• they are connected to ancient documents believed to be from the time of the
Trojan War and before;
• they appear to represent a different script than the contemporary Greek alpha-
bet (and usually cannot be read).

If we include the information that may be adduced from the attestations in Septimius’
epistula and the work of Sextus Empiricus, the following features may be added:

• they were believed to represent a language related to (similar to, yet different
from) Greek;
• they seem to refer to syllabic writing;
• they are thought to have represented fewer of the phonemic distinctions of
Greek than the Greek alphabet.

The translations ‘red’ or ‘Phoenician’ do not adequately explain these characteristics,

least of all the strong connection with the Heroic Age. It cannot be a coincidence that all
these attestations of phoinikeia grammata refer to very ancient inscriptions, and never to
contemporary texts. They must refer to an older script, which was distinctly different from
the then current Greek alphabet. As it so happens, a different writing system that fits the
above description, namely Linear B, was in use in Late Bronze Age Greece. This leaves us
with two options: either the phoinikeia grammata in these cases consistently refer to a fic-
titious ancient writing system, which coincidentally shares all historical and formal char-
acteristics of Linear B, or, applying Occam’s razor, they are in fact this Late Bronze Age
script.85 Following the proposal of Ahl, the adjective phoinikeia thus originally referred
to palm leaves, which have long been suspected to have been the primary writing material
for Linear B.86
From at least the fifth century onwards, however, it was no longer common knowledge
that the expression phoinikeia grammata referred to the Linear B script, and people like
Herodotus instead connected it to early alphabetic writing, which he thought had been
brought to Greece by the Phoenicians (hence ‘Phoenician letters’). His reinterpretation
of the phoinikeia grammata as ‘Phoenician letters’ may have been triggered by ancient
alphabetic inscriptions he saw at Thebes, combined with his admiration for the
Phoenicians.87 As a consequence, phoinikeia grammata also came to refer to (Archaic) alpha-
betic inscriptions. Some authors, however, were still aware of the term’s original meaning
and used it to refer to Linear B writing.
The above scenario has two important ramifications, which might be conceived as prob-
lematic at first glance. First of all, it implies that the Linear B script was written on palm
leaves. As mentioned above, this idea has been put forward before, but it has not found

85
See also, for example, the remark of Carpenter (1935) 8: ‘Since the locality is right, the environment is right,
why should we not accept the Greek folk memory tradition as a true tradition and, taking the Herodotean phrase,
call this Helladic writing “Cadmean letters”?’ Similarly, Phillips (1957) 272–75 suggested that the story of
Palamedes as inventor or introducer of writing must refer to Aegean writing systems and not the alphabet.
86
See, for example, Evans (1921) 638; Myres quoted in Evans (1952) 2; Diringer (1953) 42; Ahl (1967) 188. Note
that this interpretation would also explain the remark by an anonymous author in the Codex Vaticanus, in which
the first uses of phoinikeia grammata are contrasted to the later alphabet (χρὴ εἰδέναι ὅτι πρότερον οἱ Ἓλληνες
φοινικοῖς ἐχρῶντο γράμμασιν), which is opposed to the later (ὕστερον) invention of the Greek alphabet (cod. Vat.
Graec. 711, fol. 97, 14th century AD); see Devreesse (1950) 197; Edwards and Edwards (1977) 134, who apparently
take phoinikeia grammata here to refer to an early stage of the Greek alphabet, though this does not explain the
contrast with the subsequent invention of the latter.
87
For more on his possible motives, see below, section XII.

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 The Journal of Hellenic Studies 235

general acceptance. In contrast to Linear A, which is commonly believed to have been used
on perishable materials,88 there is no such consensus with respect to Linear B. Some schol-
ars assume that Linear B must have been written on perishable material,89 but others
maintain that this script was restricted to writing on the more durable clay.90
Secondly, taking phoinikeia grammata to refer to Linear B implies that in Classical
Greece there was awareness of the existence and the appearance of this Late Bronze
Age writing system, a statement which may raise some eyebrows. Since the palm leaves,
inscribed or not, are now irretrievably lost and the minds of those living in ancient times
are equally inaccessible, we have to rely on indirect evidence to substantiate these claims.
Fortunately, there are sufficient data at hand to support the scenario proposed here. Let us
first turn to the use of Linear B on perishable materials.

VII. In palmarum primo scriptitatum: evidence for palm-leaf writing from Linear
B and later sources
The word φοῖνιξ is already attested in Linear B texts, where we find the noun po-ni-ke and
the adjective po-ni-ki-jo/a.91 Unfortunately, the contexts are not very helpful in determin-
ing the meaning of this word. The information provided by the ultra-brief texts is simply
too limited to give a reliable translation. The word po-ni-ke refers several times to a deco-
rative motif on furniture, where the most plausible meaning is ‘palm tree’.92 As an adjec-
tive, it may also describe chariots and textiles, in which case it may mean ‘red’. It further
occurs in combination with some sort of spice or condiment measured by weight. Here, it
probably refers to dates from a date palm.93 No clear connection between po-ni-ke/po-ni-ki-
jo/a and writing has been attested, but this is hardly to be expected, considering the types
of document at hand. Based on the available Linear B evidence, one can only conclude that
it is highly likely that the word φοῖνιξ was already used to refer to palm trees in the
Mycenaean Age.
Though the Linear B texts may not offer direct evidence for the use of palm leaves for
writing, they do present indirect evidence for the use of perishable writing materials. First
of all, there is the format of the surviving Linear B texts on clay. As has long been pointed
out, the shape of the most common type of Linear B tablets resembles the shape of palm
leaves (see fig. 1). This tablet type has therefore been dubbed the ‘palm-leaf tablet’.94 The
choice of this tablet shape was deliberate; clay is a malleable substance and can be formed
into all kinds of shapes and sizes (and other forms are also attested). An obvious explana-
tion for why the clay was shaped into this form would be that it was imitating an already
existing type of document, namely one written on palm leaves.95

88
See, for example, Weingarten (1983); Krzyszkowska (2005) 155–57; Hallager (1996) 135–45; Perna (2017) 72–76.
89
See above n.86 and more recently, for example, Driessen (2000) 186–87; Palaima (2003) 171–72; Palaima
(2011); Waal (2021).
90
See, for example, Bennet (2001) 27–28; Perna (2011) 18–19; Steele (2017) 154 n.5.
91
See Chadwick and Baumbach (1963) 254 s.v. φοῖνιξ; Ventris and Chadwick (1973) 573; Ruijgh (1967) 168; Aura
Jorro and Adrados (1993) 138.
92
The translation ‘griffin’ has also been proposed; see Aura Jorro and Adrados (1993) 138. But the meaning
‘palm tree’ is more likely; see Ventris and Chadwick (1973) 502 and Palmer (1957) 89.
93
Melena (1975).
94
Note that this tablet shape is thus far unattested in the Linear A corpus.
95
Here, a brief note regarding the term ‘palm leaf’ in relation to writing seems in order. Palm-leaf manuscripts
usually refer to manuscripts from South and Southeast Asia, which are written on dried or smoke-treated palm
leaves. These are made from different types of palm tree than those that grow in certain parts of the
Mediterranean, where the date palm is most common. When date palms served as a writing material, the rib
or spine of the leaf were often used (see, for example, fig. 2), and this may well have been the case in
Mycenaean Greece as well, though it should not be excluded that they also wrote on the leaves. When cut into

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 236 Willemijn Waal

Fig. 1. Palm-leaf tablet from Pylos (PY Eb 1176), Courtesy of The Pylos Tablets Digital Project and The Palace of
Nestor Excavations, The Department of Classics, University of Cincinnati.

Further, it has often been noted that the complex and curved forms of the Linear B
characters are better suited to be written in ink with pen or brush than to be incised
in the coarse material clay.96 Alternatively, no ink was used, but palm leaves or ribs were
incised with a stylus. As examples from South Asia show, this technique is very suitable for
making round letters.97 It is telling that the Linear B sign forms are retained over time
without any simplification or abstraction, which would have made writing in clay consid-
erably easier. By contrast, we do very clearly see such developments in the cuneiform
script, which was written almost exclusively on clay. The most logical explanation for
the unchanging complexity of the Linear B sign forms is that clay was not the primary
writing material for this script, that it was mostly used on softer materials, such as palm
leaves.
Another indicator for the use of Linear B writing on perishable material is the awkward
ratio of the relatively high number of individuals who were involved in writing to the
relatively few and short extant records. This imbalance would be explained if one assumes
that writing was used on a much larger scale on perishable materials.98 This would also
account for the limited scope of these tablets. The surviving clay records only deal with
local economic administration; texts of other genres are completely absent. However, con-
temporary Hittite texts indicate that the Aegean world participated in international dip-
lomatic correspondence, which suggests that a more diverse textual corpus must have
existed.99 Whether they made use of their own writing system and/or the cuneiform script
for this purpose is open to debate, but regardless of the type of script that was employed, it
implies a wider use of writing than for the sole purpose of recording of local palatial
administration.100 Needless to say, the perishable materials used for writing were not

smaller pieces, the palm ribs have a layout that is somewhat comparable to narrow palm leaflets, and the sur-
viving Linear B clay tablets could in principle mimic both. For the sake of convenience, I will adhere to the con-
ventional term ‘palm-leaf tablet’.
96
See, for example, Evans (1921) 638; Palaima (2003) 171.
97
See, for example, Padmakumar et al. (2003) 128; Mahadevan (2003) 178, who, with respect to the Tamil-
Brāhmī script, traces the tendency to convert angular and rectilinear letters and medial vowel signs into cursive
shapes (ultimately resulting in the Vatṭeluttu, literally ‘rounded script’) to the practice of writing on palm leaves
with an iron stylus.
98
A different explanation has been proposed by Bennet, who suggests that these scribes were in fact admin-
istrators at the highest level, presumably members of the elite, who spent much of their time supervising activi-
ties (Bennet (2001) 31–35). Though this scenario is certainly possible, it is not without problems; see Palaima
(2003) 176–77.
99
See, for example, Melchert (forthcoming), Beckman et al. (2011) 138–39; Hoffner (2009) 299; Surenhagen
(2008) 260–65; Bryce (2003) 199–200; Waal (2021) 213–14.
100
The argument that the Linear B script would not be suitable for the composition of longer, more complex
texts is not valid, as comparable writing systems elsewhere show. It should further be borne in mind that the

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 The Journal of Hellenic Studies 237

Fig. 2. Palm midrib with incised alphabetic inscription from Yemen (L024), 11th-10th c. BC, Courtesy of Stichting
Oosters Instituut, Leiden University Libraries. Photograph: Wim Vreeburg.

necessarily restricted to palm leaves; perhaps they also made use of other ephemeral materi-
als, such as leather, parchment, papyrus or wood, for more elaborate compositions.101
From a global perspective, the choice of leaves, as well as other parts of trees such as
wood or bark, as a primary writing material is nothing exceptional (fig. 2). Trees have
been, and still are, a very common source of writing materials in many regions of the
world.102 The popularity of leaves is reflected in today’s terminology for script bearers
in many languages (for example, folia, Blatt, hoja, feuille, ‘leaf’, etc.). When the Linear B
scribes, who were accustomed to writing on palm leaves, happened to write on clay, they
stuck to the same scribal conventions, including the shape of the documents.103
Confirmation of the practice of writing on palm leaves is found in later sources, such as
Photius and the Suda.104 It is also mentioned by Pliny the Elder (23–79 AD) in his Historia
naturalis. Pliny (HN 13.21), citing Varro, relates that the use of papyrus for writing was
invented in Alexandria, and that before that time, people wrote on other writing materials,
including palm leaves. Varro informs us that paper owes its discovery to the victorious
career of Alexander the Great, and his founding of Alexandria in Egypt:

antea non fuisse chartarum usum: in palmarum foliis primo scriptitatum, dein quar-
undam arborum libris. postea publica monumenta plumbeis uoluminibus, mox et
priuata linteis confici coepta aut ceris; pugillarium enim usum fuisse etiam ante troi-
ana tempora inuenimus apud Ηomerum.
[B]efore that period paper had not been used, the leaves of palms having been
employed for writing first, and after that the bark of certain trees. In succeeding peri-
ods, public documents were inscribed on sheets of lead, soon private memoranda

abbreviated and restricted nature of the surviving Linear B texts does not provide us with a full picture of the
script (or the language).
101
For evidence of the use of wooden diptychs, see Shear (1998).
102
See, for example, Diringer (1953) 37–44; Padmakumar et al. (2003).
103
This implies that the surviving Linear B (and A) clay tablets were rather the exception than the rule. The
reason why some documents were written down on clay and not others is unclear; see Waal (2021) 219–20.
104
See above, section II. Note that the ‘palm leaf’ interpretation of phoinikeia grammata is usually seen as a
device born out of nationalistic concerns to retain the invention on Greek soil; see Edwards and Edwards
(1974) 52; Jeffery (1967) 157–58; Ceccarelli (2013) 68. Edwards and Edwards (1974) 52 consider it telling that writing
on palm leaves is not mentioned at Htd. 4.43 or Theophr. Hist. pl. 4.2.7, 9.4.4, while other uses of the palm in
various parts of the world are mentioned. However, Theophrastus discusses a special type of palm tree (the doum
palm) that grows in Egypt. Since this entire section deals with Egypt, one can hardly expect references to palm-
leaf writing in the Aegean. The same applies to the account of Herodotus, which tells the story of Sataspes, who,
when attempting to circumnavigate Africa, apparently saw short people along the coast of distant regions wear-
ing clothes made of palm leaves.

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 238 Willemijn Waal

were also impressed upon linen cloths, or else engraved on tablets of wax; indeed, we
find it stated in Homer, that tablets were employed for this purpose even before the
time of the Trojan War.

Though Pliny’s sense of chronology leaves much to be desired (the invention of papyrus
of course occurred long before Alexander the Great), the existence of the writing materials
that he lists is confirmed by archaeological evidence and/or other sources. With respect to
the practice of writing on leaves, further evidence is provided by Diodorus Siculus. This
historian not only informs us about the grammata phoinikeia on the lebēs of Cadmus (see
above, section VI), but also about the use of leaves (πέταλα) as writing material, in this case
in Sicily. In book 11, he mentions in passing the phenomenon of petalism (πεταλισμός) in
Syracuse, a banishment practice similar to ostracism in Athens (11.87.1–3):

Now among the Athenians each citizen was required to write on a potsherd (ostrakon)
the name of the men who, in his opinion, was most capable through his influence of
tyrannizing his fellow citizens; but among the Syracusans the name of the most influ-
ential citizen had to be written on an olive leaf (πέταλον ἐλαίας), and when the leaves
were counted, the man who received the largest number of leaves had to go into exile
for five years . .. Now while the Athenians called this kind of legislation ostracism,
from the way it was done, the Syracusans used the name petalism (πεταλισμόν).105

Olive leaves are obviously quite small, but since only a name needed to be jotted down,
they would have sufficed for this particular purpose. Though referring to a later period,
this attestation is a meaningful example of the use of leaf writing in the Greek-speaking
world, as already noted by Ahl.106

VIII. Greek awareness of their written past

The interpretation of phoinikeia grammata suggested above implies that in Classical times,
at least in some circles, there was awareness of the existence of Linear B, and possibly
other writing systems in the Aegean in the past. One only has to think of Arthur
Evans’ observation that as late as the early 20th century Cretan women were wearing
Linear B tablets that they had found as charms, to realize that in Classical times similar
discoveries of Late Bronze Age writing were undoubtedly made.107 There is in fact textual
evidence to support this claim. The fabricated story of the rediscovery of the Journal of
Dictys discussed above (section IV) appears to have been based on actual discoveries.108 A
somewhat similar report relates the discovery of an ancient document in the tomb of
Alcmene, the mother of Heracles, of which Alain Schnapp has remarked that ‘it does
not take too much imagination for today’s archaeologists to recognize a Mycenaean
burial’.109 When the grave was opened, a bronze tablet with a long inscription in an
unknown script was revealed. A detailed account is given in the Moralia of Plutarch
(46–120 AD). In the following passage, Phidolaus writes about the opening of the grave:

In front of the monument there lay a bronze tablet full of strange and certainly very
ancient letters (γράμματα πολλὰ θαυμαστὰ ὡς παμπάλαια), for nobody could make
anything of them, even though once the bronze was washed they were very clearly

105
Translation based on Oldfather (1956) 347–49.
106
Ahl (1967) 194 n.20.
107
Cf. Ahl (1967) 189.
108
See, for example, Evans (1909) 108; Astour quoted in Owen and Young (1997) 33.
109
Schnapp (1996), 55.

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 The Journal of Hellenic Studies 239

visible; their form was particular and strange, resembling very much that of Egyptian
characters. Agesilaus accordingly, as the story goes, sent copies to the king (of Egypt),
requesting him to show them to the priests to see if they could interpret them.110

It has been suggested that the mysterious signs on this tablet, which look like Egyptian
hieroglyphs, correspond to Linear B.111 Yet another mention of the discovery of ancient
documents are the bronze tablets dug up by the father of the historian Acusilaus of Argos,
which supposedly served as the source for his Genealogies.112 Further references to ancient
inscribed objects include a remark in Ampelius’ Liber Memorialis (8.5) about litterae Palamedis
that were deposited in the temple of Apollo at Sicyon together with other ancient heir-
looms, such as the shield of Agamemnon,113 and a pillar with ancient characters in the
Mirabilia of Pseudo-Aristotle.114 Though such discovery stories, like the Fundbericht of
the Diary of Dictys discussed above, may reflect a literary topos rather than historical real-
ity, and the ancient finds described may not be ‘real’ in the sense that it was not the actual
tomb of Alcmene that was unearthed and the aforementioned shield did not really belong
to Agamemnon,115 they could nonetheless be based on genuine discoveries of objects and
remains of earlier times. The ancient Greeks were regularly confronted with the physical
reality of their (heroic) past, in the form of tombs, ruins, pictures and ancient artefacts.116
Some of these remains were, rightly or wrongly, used as evidence to prove the historicity
of Homer’s epics and the Trojan War, showing how much importance the Greeks (and
Romans) attached to seeking out their glorious past.117 The value attributed to relics of
the Heroic Age was such that it even led to the creation of forgeries.118
Apart from physical discoveries of Late Bronze Age writing, there are also numerous lit-
erary references to writing in the heroic past. As shown above (section IV), phoinikeia gram-
mata appear several times in the context of the Trojan War and before. Apart from the
examples discussed above, there are more literary texts, especially tragedies, that make
mention of the use of writing in this period.119 In these cases, however, it is not stated that

110
Plut. De gen. 577, translation based on Hani (1980) 77–78. For discussion of this passage, see Parker (2010).
111
See, for example, Evans (1909) 107–08; Forsdyke (1956) 43; Larson (1995) 92; Schwartz (1958) 81; Schachter
(1981) 14; Schnapp (1996) 55; Astour quoted in Owen and Young (1997) 33. Note that later on in the text another
ancient document is mentioned, which contained characters in the form of the script current in the time of king
Proteus, which Heracles had learned. It was supposedly deciphered by Chonuphis of Memphis (Plut. De gen. 578).
112
Suda α 942 Adler = T 1, Fowler (2000) 1. See also Jeffery (1967) 159 n.27; Higbie (1999) 55; Fowler (2013) 624;
Andolfi (2019) 13–15.
113
See Jeffery (1967) 159 n.27. As shown by Scheer (1996), the inventory of this temple given by Ampelius is
quite similar to those known from other Greek cities. The Palamedis litterae are probably to be understood as a
variant of ‘Cadmeian letters’, since Palamedes was credited with the invention of writing by some traditions. The
Palamedis litterae may thus also refer to Linear B (or A). Alternatively, however, they may in this case refer to the
famous letter falsified by Odysseus leading to the death of Palamedes; see Scheer (1996) 368.
114
See also below, section X.
115
Cf. Scheer (1996) 361–62.
116
Boardman (2002) especially 9–14; Schnapp (1996), especially 45–56; Higbie (2017).
117
Cf. Boardman (2002) 191.
118
Higbie (2017) especially 209–23.
119
See Easterling (1985) 3–6. For example, Euripides’ Hippolytus presents literacy in the period before the Trojan
War, in the lying letter of Phaedra and the names written on sails; see also S. West (1985) 294 n.77. Oeax, the
brother of Palamedes, tries to send a message to their father Nauplius by inscribing oar blades and throwing
them into the sea; see also Scodel (1980) 58; Ceccarelli (2013) 82–84. In Hyginus’ Fabula 105, Odysseus fabricates
a letter to Priam allegedly written by Palamedes, causing the downfall of the latter; see, for example, Ceccarelli
(2013) 78. Likewise, in Iphigenia in Tauris, Iphigenia writes a letter, as does Agamemnon in Iphigenia in Aulis. In
Aeschylus’ Supplices, Pelasgus explicitly mentions written tablets (πιναξίν ἐγγεγραμμένα) and sealed papyri docu-
ments (ἐν πτυχαῖς βύβλων κατεσφαγισμένα (946–47). Sophocles refers to an ancient tablet with writing in the
Trachiniae (157: παλαιὰν δέλτον ἐγγεγραμμένην) and portrays Heracles as literate when he writes down prophe-
cies from dictation at Dodona (1166).

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 240 Willemijn Waal

this was done in phoinikeia grammata and in some cases it is evident that not Linear B, but
alphabetic writing is implied.120 When reviewing the numerous references to writing in the
Heroic Age, Patricia Easterling rightly observes that writing in the distant past was appar-
ently not seen as anachronistic. She draws attention to the striking contrast between the
frequent use of writing in the Heroic Age in tragedy and the Homeric model, in which writ-
ing does not play a vital part.121 The alleged opposition between the tragedians and Homer
with respect to allusions to writing in the Heroic Age may not be as significant as it appears,
however, and may even be non-existent. References to writing in Homer may be scarce, but
they are not entirely absent. The most famous and unambiguous example is the episode in
Il.6 about the hero Bellerophon, who unknowingly carries and delivers his own death war-
rant written by his father-in-law Proitus. As argued by Jenny Strauss Clay, the episode in Il.
7.87–91, when Hector imagines an epitaph to his glory, also implies a knowledge of writing.
Further, there is a possible reference to writing in Il. 7.175–89.122
Unfortunately, the significance of the seemingly limited role of writing in the Iliad has
been grossly overstated and misinterpreted. It has been taken as evidence that Homer
himself was illiterate, or that writing was deliberately suppressed in the epic, whereas
the most obvious explanation is much more prosaic: writing does not feature prominently
because this skill is not particularly pertinent to an heroic epic about war. The observation
made by Nathaniel Schmidt over a century ago has lost none of its relevance:

A careful perusal of works such as Apollonius’ Argonautica, Vergil’s Aeneid, Lucian’s

Pharsalia, Silius’ Punica, Statius’ Thebaid, Tasso’s Gerusalemme Liberata, Camoens’
Lusiadas, Milton’s Paradise Lost, the Mahabharata, the Ramayana, the Kalevale, with a
view to discovering allusions to writing, brings home to the conviction that epic
poets, excogitating their verses, pen in hand, very rarely think of mentioning the
art they so constantly practice.123

The fact that there are few references to writing in the Iliad is thus not necessarily due
to conscious avoidance or downplaying, nor to the fact that Homer himself and/or the
world he wanted to describe was illiterate, but simply because there was no need or moti-
vation to refer to this mundane activity, except in the case of the Bellerophon episode,
where the written message constitutes a crucial element of the story. The limited role
of writing in Homeric epic is just as insignificant as the fact that no cats are mentioned
in the Bible, and their nonappearance can hardly be used as evidence for the absence of
script, or domestic felines, in certain time periods.

IX. σήματα versus γράμματα

Though the references to writing in Homer may be scant, the terminology used is of great
significance. In the Bellerophon passage in Iliad book 6 just mentioned, Proitus writes down
many ‘baleful signs’ (6.169: σήματα λυγρὰ) on a tablet (6.170: γράψας ἐν πίνακι).
Bellerophon is asked to show the message (6.176: σῆμα) to his host. Later on, this message
is described as ‘horrible᾽ (6.178: σῆμα κακὸν). Homer consistently uses the word σῆμα

120
In the lost Theseus of Euripides, a herdsman describes the letters forming the name Theseus (*fr. 382Kn Jouan
and Van Looy (2000) 158–59; Collard and Cropp (2008a) 420–21). Further, in the version attested in a scholion on
Euripides’ Orestes, Odysseus forces a Trojan prisoner to write a letter in Phrygian letters.
121
Easterling (1985) 5.
122
Strauss-Clay (2016). See below, section IX.
123
Schmidt (1920) 66.

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 The Journal of Hellenic Studies 241

rather than expected γράμμα or στοιχεῖοv to refer to writing.124 The word σῆμα also
occurs in another possible reference to writing in book 7.125 The scene runs as follows.
When the heroes have to decide who is to be the one to fight Hector (7.175–89), each
man marks his lot (κλῆρον ἐσημήναντο ἕκαστος) and casts it into the helmet of
Agamemnon. Nestor subsequently shakes the helmet and the lot of Ajax leaps out.
Next, the herald walks around in the crowd, showing the lot to all the chieftains of the
Achaians. None of them recognizes it (7.185: οἳ δ᾽ οὐ γιγνώσκοντες ἀπηνήναντο
ἕκαστος). When the herald finally reaches Ajax, he places the lot in his hand, whereupon
Ajax recognizes the token as his own and rejoices (7.189: [Aἴας] γνῶ δὲ κλήρου σῆμα ἰδών,
γήθησε δὲ θυμῷ).
This passage can be, and has been, interpreted in various ways.126 It could be the case
that the heroes each marked their lot with a personal sign that was unrelated to writing
and that was meaningful only to themselves, or perhaps with an impression of their per-
sonal seal. In this case, none of the heroes, except for Ajax, would have known to whom the
winning lot belonged, unless of course they were familiar with each other’s personal signs
or seals. Alternatively, the heroes do make use of writing, each jotting down (part of) their
name. In this interpretation, everyone who sees the lot instantly understands that it
belongs to Ajax, and the remark that they do not recognize it (οὐ γιγνώσκοντες) would
indicate that they do not recognize it as their own. The lot is thus not shown to all in
search of its rightful owner, but so that everyone can see for himself to whom it belongs
and that there has been no foul play. Accordingly, the fact that it is first shown to all the
others, and only then handed over to Ajax, is not a coincidence. An interesting parallel
from later times is provided by the practice of inscribing with a letter lots, which were
also shown around for inspection.127
Though the second example may be ambiguous, the σήματα of the message of Proitus in
book 6 without doubt refer to writing. In section V it has been suggested that the use of the
word σῆμα by Timon refers to Linear B writing. In light of the evidence presented above,
this is also the most obvious and logical interpretation of the signs foretelling
Bellerophon’s death.
According to LSJ, the word γράμμα, ‘that which is drawn’, can refer to letters and mes-
sages, but also to pictures and marks. The principal meanings of σῆμα are ‘mark’, ‘omen’ or
‘token’. It can, for instance, be used for a mark on an animal’s head, signs from heaven,
heavenly bodies, or a sign by which a grave is known.128 If one accepts that σῆμα in the
above cases refers to Linear B, its usage would be comparable to our modern use of the
word ‘sign’ (Zeichen, teken, signe, etc.) to denote the elements of pictorial, logosyllabic writ-
ing systems, such as Linear B. Like σῆμα, γράμμα can also refer to marks, signs or pictures,
but, unlike σῆμα, it can in addition refer to letters, that is, the elements of the alphabet. In
other words, γράμμα can refer to both logographic/syllabic and alphabetic writing sys-
tems, and σῆμα only to the former.

124
LSJ proposes ‘written characters or symbols’ for Il. 6.168. Note that LSJ’s interpretation ‘token by which any
one’s identity or commission was certified’ is not attractive with respect to Il. 6.176 and Il. 6.178, since Bellerophon
is only asked to show his σῆμα on the tenth day. The fact that this σῆμα is described as horrible (κακὸν) in Il. 6.178
confirms that the fatal message written by Proitus described at Il. 6.168 (σήματα λυγρὰ) is meant here. For a
discussion of this passage, see, for example, Steiner (1994) 15–16 and also Ford (1992) 132, who holds that
Oriental writing is meant here. For the use of the term στοιχεῖον to refer to letters, see LSJ s.v. στοιχεῖον II.
125
Interestingly, the word σῆμα also occurs in the likely reference to writing in Il. 7.87–91 (see above), where it
refers to the mark of the tomb of Hector’s opponent.
126
For discussion, see also Ford (1992) 138–39 and Steiner (1994) 13–15.
127
See LSJ s.v. γράμμα II.4 with references. Besides γράμμα, the word στοιχεῖοv may also be used in this context
(Arist. [Ath. Pol.] 64.4).
128
On the various uses of σῆμα, see also Steiner (1994) 10–40 and, with respect to the use of signs in Homer,
Ford (1992) 131–71.

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 242 Willemijn Waal

X. Rediscovery or continued tradition?

Based on the available material, it cannot be established with certainty whether the allu-
sions to writing in the Late Bronze Age are to be seen as later projections triggered by
discoveries of ancient texts in Classical times, or as reflections of a continued tradition
of ancient legends in which writing played a part. In light of the growing evidence pointing
to continuity between the Late Bronze Age and the Early Iron Age, and the fact that Greek
historians themselves saw no breach between the Heroic Age and their own era,129 I would
certainly not exclude the latter, but the former scenario might be more palatable to most.
Obviously, the two scenarios are not mutually exclusive, but could rather reinforce
each other.
Perhaps more controversial is the question of whether these ancient texts could still be
read and understood. This idea was radically dismissed by John Forsdyke, who stated that
all knowledge of Minoan and Mycenaean scripts was completely lost and claimed that ‘it is
also certain that the historical Greeks could not read nor even recognise the prehistoric
scripts’.130 As already remarked by Hagen Biesantz, who holds a more nuanced and positive
view, however, Forsdyke’s claim cannot be substantiated, and there are no grounds for
assuming that no knowledge of the script of the ancestors had been preserved anywhere
by anybody.131
Indeed, though it may well be too optimistic to assume that actual knowledge of Linear
B survived, it is also unfair to categorically deny all ancient Greeks any historical under-
standing of their written past. Of particular significance is the account in the Mirabilia
(Pseudo-Aristotle) about an ancient pillar with ancient characters, which is brought to
the Ismenium at Thebes, already briefly referred to above:

In the country called Aeniac, in that part called Hypate, an ancient pillar is said to
have been found; as it bore an inscription in archaic characters (ἐπιγραφὴν ἀρχαίοις
γράμμασιν) of which the Aenianes wished to know the origin, they sent messengers to
Athens to take it there. But as they were travelling through Boeotia, and discussing
their journey from home with some strangers, it is said that they were escorted into
the so-called Ismenium in Thebes. For they were told that the inscription was most
likely to be deciphered there, as they possessed certain offerings bearing ancient let-
ters similar in form. There, having discovered what they were seeking from the
known letters, they transcribed the following lines.132

It is intriguing that the ancient inscribed pillar mentioned is eventually brought to the
Ismenium of Thebes to be deciphered, the same location where Herodotus claims to have
seen inscriptions in ‘Cadmeian letters’ and, much later, Pausanias (ca. 110–180 AD) report-
edly saw an ancient, presumably inscribed, tripod dedicated to Heracles.133 It is a tantaliz-
ing thought that this sanctuary not only housed ancient inscribed objects, but also experts
with specialized knowledge of these inscriptions.134 Due to the lack of more solid evidence
(the Mirabilia may indeed not be the most reliable of sources), however, this claim cannot
be substantiated. Likewise, the reference to the translation of the ancient bronze tablet
mentioned in Plutarch’s Moralia by specialists in Egypt should be taken cum grano salis.135

129
Cf. Bintliff (2012) 211.
130
Forsdyke (1956) 40.
131
Biesantz (1958) 56.
132
Ps.-Arist. Mir. 133 (843b), translation based on Hett (1936) 305–07.
133
Paus. 9.10.4.
134
From the fact that the pillar was initially taken to Athens, one might infer that this city also had a reputation
for housing such expertise.
135
Plut. De gen. 578f–79.

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 The Journal of Hellenic Studies 243

For the moment, one can only cautiously conclude that some strata of the population at
least knew of the existence of Linear B (and possibly other Late Bronze Age writing sys-
tems), and that in some highly specialized circles, such as the Ismenium at Thebes, rudi-
mentary knowledge about its basic characteristics was present. As already mentioned
above (section V), the continued existence of a syllabic writing tradition at Cyprus, which
was a derivative of Linear A, makes it conceivable that there was awareness, or at least
suspicion, of the fact that the Linear B script was (logo)syllabic rather than alphabetic.
This distinction is made explicit in the terminology used by authors like Homer and
Timon (apud Sextus Empiricus), who referred to Linear B as ‘signs’ (σήματα) rather than
‘letters’ (γράμματα).
Either way, be it the result of later projections or of continued tradition, to the minds of
the ancient Greeks the Heroic Age was a literate period. This is clear from the references to
writing in tragedies situated in the Trojan War period, as well as from the fact that all
narratives about the invention of writing are placed in the Heroic Age. In some cases
(though certainly not all), the kind of writing that was used in ancient times is specified,
and referred to as phoinikeia grammata (or Pelasgic/Palamedian/Cadmeian letters), imply-
ing a dissimilarity with the contemporary writing system. This is all hardly surprising,
since a different writing system was in use in the Late Bronze Age, and samples of this
ancient script did survive and were known in later times.136

XI. Pelasgic letters

The interpretation of phoinikeia grammata suggested here may also shed light on the
intriguing expression ‘Pelasgian’ or ‘Pelasgic’ letters. The term ‘Pelasgians’ (Πελασγοί)
was used by classical authors to refer to the ancient inhabitants of Greece, in the times
before the Trojan War.137 The term ‘Pelasgic letters’ first appears in Dionysius
Scytobrachion (third century BC), whose works included a history of the Trojan War.138
His account of the development of the Greek alphabet is quoted by Diodorus and runs
as follows:

Φησὶ τοίνυν παρ’ Ἕλλησι πρῶτον εὑρετὴν γενέσθαι Λίνον ῥυθμῶν καὶ μέλους, ἔτι δὲ
Κάδμου κομίσαντος ἐκ Φοινίκης τὰ καλούμενα γράμματα πρῶτον εἰς τὴν Ἑλληνικὴν
μεταθεῖναι διάλεκτον, καὶ τὰς προσηγορίας ἑκάστῳ τάξαι καὶ τοὺς χαρακτῆρας
διατυπῶσαι. κοινῇ μὲν οὖν τὰ γράμματα Φοινικήια κληθῆναι διὰ τὸ παρὰ τοὺς
Ἕλληνας ἐκ Φοινίκων μετενεχθῆναι, ἰδίᾳ δὲ τῶν Πελασγῶν πρώτων χρησαμένων
τοῖς μετατεθεῖσι χαρακτῆρσι Πελασγικὰ προσαγορευθῆναι.
He says then that among the Greeks Linus was the first to discover the rhythms and
song, and when Cadmus brought from Phoenicia the letters, as they are called, Linus
was again the first to transfer them into the Greek language, to give a name to each
character, and to fix its shape. Now the letters, are officially called ‘Phoenician’
because they were brought to the Greeks from the Phoenicians, but unofficially,
because the Pelasgians were the first to make use of the transferred characters, they
were called ‘Pelasgic’.139

136
Note that the idea of early literacy in Greece was also viewed with scepticism by some ancient authors, such
as, for example Josephus (first century AD); see Jeffery (1967) 161 with references.
137
Gschnitzer (2006).
138
Ní Mheallaigh (2012) 189.
139
Dionysius Scytobrachion, FGrH/BNJ 32 F 8 (= Diod. Sic. 3.67.1), translation largely based on Oldfather (1953)
305.

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 244 Willemijn Waal

It is unclear how one should understand κοινῇ and ἰδίᾳ in this context and several dif-
ferent proposals have been made. The Loeb translation of Charles Henry Oldfather (fol-
lowed by Paola Ceccarelli) suggests that the letters ‘as a group’ are called ‘Phoenician’
and ‘as single letters’ are called Pelasgic.140 Ní Mheallaigh prefers ‘commonly’ vs ‘pri-
vately’,141 and Jeffery translates ‘officially’ and ‘unofficially’ (the interpretation chosen
here).142 An exhaustive discussion is given by Aldo Corcella, who suggests that to
Dionysius the ‘Phoenician’ letters are the alphabet in general, and the ‘Pelasgic’ ones relate
to the first ‘Greek’ alphabet which was created by Linus based on the Phoenician alphabet
introduced by Cadmus, thus reconciling different traditions.143 Creative as these interpre-
tations may be, none of them yields a satisfactory result and the account of Dionysius fails
to persuade. Jeffery convincingly concludes that his ‘ill-fitting explanation’ shows that the
phenomenon of ‘Pelasgic letters’ was not an invention of Dionysius himself, but that he is
rather struggling with two already existing terms.144
The quandary vexing Dionysius would be easily solved, however, if one took the phoi-
nikeia grammata and Pelasgic letters to refer to Linear B writing. The most common way
(κοινῇ) to refer to this script was connected to its primary writing material, palm leaves
(phoinikeia grammata). In addition, there was a less frequently used term (ἰδίᾳ) that referred
to the users of this script, the Pelasgians. This term was probably not used by the users of
Linear B themselves, but rather a later designation.145 In Classical times, both of these
terms were in use by the Greeks to refer to the script of their ancestors, thus distinguishing
this ancient writing system from their own contemporary alphabetic writing, to which
they simply referred as grammata (or stoicheia). Over time, however, the original meaning
of phoinikeia grammata came to be misunderstood (as explained above) and the expression
was linked to alphabetic writing. A similar fate befell the expression ‘Pelasgic writing’,
undoubtedly because the existence of Linear B writing was no longer common knowledge.
Subsequently, one was faced with the awkward situation that two very different, mutually
exclusive ethnika, ‘Phoenician’ and ‘Pelasgic’, came to refer to the same script, the Greek
alphabet, a riddle that Dionysius failed to resolve convincingly. Ní Mheallaigh concludes
that ‘Dionysius’ imaginary palaio-literary landscape was ... generously littered with Greek
texts written in these so-called “Phoenician” or “Pelasgic letters” which the scholar seems
to have taken some trouble to accommodate to the Greek alphabet’.146 In the scenario pro-
posed here there is nothing fictional about this literary landscape; the terms ‘Phoenician’
and ‘Pelasgic’ letters reflected the memory of the use of Linear B in the Late Bronze Age,
but this was no longer recognized by Dionysius and many of his colleagues.

XII. ‘On the malice of Herodotus’

The reliability, or lack thereof, of Herodotus has been a much-discussed topic since
antiquity.147 As demonstrated above (section I), there is ample reason to question his
account of the introduction of the alphabet. Apart from the obvious chronological
difficulties, the explanation offered by Herodotus is also not particularly cogent from a

140
Oldfather (1953) 305; Ceccarelli (2013) 362.
141
Ní Mheallaigh (2012) 190.
142
Jeffery (1967) 160.
143
Corcella (1986).
144
Jeffery (1967) 160.
145
Alternatively, one could take the Pelasgic letters to refer to Linear A, and phoinikeia grammata to Linear B, but
this has to remain a conjecture.
146
Ní Mheallaigh (2012) 190.
147
For a discussion of the criticisms of his work in antiquity, see, for example, Evans (1968) 11.

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 The Journal of Hellenic Studies 245

semantic perspective. When taking over a writing system, it is not uncommon that the
recipient party borrows the terminology used for writing in the source language. If we
think of the ancient Near East, for example, we see that the Sumerian terminology for
‘tablet’ and ‘scribe’ are taken over by Akkadian speakers who adopted this script.
Likewise, the Chinese term hanzi was taken over in Japanese (kanji) and Korean (hanja)
together with the Chinese writing system. It is, however, less self-evident that the writing
system would be named after the people who introduced it.148 The use of the ethnikon
‘Phoenician’ would make sense if the writing system were contrasted to an already existing
one (compare, for example, the modern use of ‘Arabic’ versus ‘Roman’ numerals), but this
does not seem to be applicable here. Alternatively, ethnika may be used to designate certain
(erotic) acts or customs that clash with, or are different from, local rules, a practice that is
still common today (for example, ‘French kissing’, ‘Russian style’, ‘going Dutch’). In
Classical Greek, the verb φοινικίζω, ‘to be like a Phoenician’ referred to (homo?)sexual
behaviour that was deemed inappropriate.149 For an activity like writing (if not juxtaposed
to a local system!), however, such an ethnic label would be highly exceptional. If one
accepts Herodotus’ Phoenician interpretation, the implication is that the Greeks initially
saw alphabetic writing as something highly outlandish and exotic. This is scarcely credible,
considering their long-standing contacts with literate peoples, as well as their own writing
traditions of more than half a millennium. The interpretation proposed here is much more
plausible; it is not exceptional that writing systems are named after the (primary) writing
material. A nice parallel is provided by the Lontara script, whose name is derived from the
Malay word for palmyra palm (lontar), of which the leaves were used for writing.
The explanation of Herodotus should first and foremost be seen as a product of his time,
and valued as such. His interpretation of phoinikeia grammata is reminiscent of his obser-
vations about ‘correct naming’. In a number of instances Herodotus mentions specifically
whether a certain name is ‘correct’ (ὀρθῶς) or not. He considers a name to be correct when
it signifies something important and accurate about the essence of the object or the indi-
vidual.150 Herodotus was not unique in this respect; correct naming was a fashionable
interest in the fifth century. With respect to the ‘Phoenician letters’, Herodotus states that
it is only ‘right’ that the Greeks named their alphabet after the Phoenicians (ὥσπερ καὶ τὸ
δίκαιον ἔφερε), as they were the ones who introduced it to Greece.
As shown by Gomme, however, Herodotus and other logographers were prone to ‘cor-
rect’ certain traditions in light of their own theories and research abroad.151 Herodotus had
a clear preference for assuming foreign, usually Egyptian, origins and the Phoenicians
often acted as intermediaries. He evidently admired the Phoenicians, which becomes
apparent in the story about the digging of the Athos canal (7.23), where he praises their
superior skills. A theory about the Phoenicians as bringers of the alphabet would certainly
have appealed to him. Whether this theory was his own creation or was already in
existence, to Herodotus’ mind, the best explanation for the expression phoinikeia grammata
was that they were ‘Phoenician letters’, referring to the Phoenician origins of the Greek
alphabet. Confirmation of the correctness of this view was easily found in the fact that the
Greek and Phoenician alphabets were clearly related, as well as in the ubiquitous
Phoenician presence throughout the Mediterranean in the fifth century BC, which
undoubtedly strengthened Herodotus’ conviction that they must have played a crucial rule
in the spreading of the alphabet. Seen in this context, the ‘Phoenician’ reanalysis of the

148
Cf. Chantraine (1972); see also the Appendix.
149
‘When it comes to shameful acts, we are more disgusted by those who act like Phoenicians than like people
from Lesbos᾽ (Galen, On the Qualities and Powers of Herbs 12.249; cf. Luc. Pseudol. 28).
150
Thomas (2000) 230; Vignolo Munson (2005) 41–43.
151
Gomme (1913a; 1913b); see also Edwards (1979) 66 and below, section XIV.

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 246 Willemijn Waal

adjective phoinikeia seems only logical, and explains its popularity. Quite possibly, the
process of ‘Phoenicianization’ did not stop at the introduction of writing, but also affected
the origins of the hero Cadmus.

XIII. Cadmus the Mycenaean or the Phoenician?

The role of the legendary Cadmus in the introduction of writing to Greece is much dis-
cussed.152 The genealogy of this hero is not straightforward and many variants existed.153
From the fifth century onwards, Phoenicia is mostly said to be his country of origin,
though some accounts connect him to Egypt. Simultaneously, he was also generally
believed to be a descendant of the Argive heroine Io.154 Interestingly, there is no explicit
mention of his foreign origins before the fifth century; in the earliest sources Cadmus is
portrayed as a typical ‘Greek’ hero.155 In modern scholarship his Oriental roots have there-
fore been questioned.156 One of the most influential scholars to do so was Gomme, who,
after examining the chronological development of the story of Cadmus from the literary
sources, suggested that the Phoenician origins of Cadmus were not part of ancient tradi-
tion, but originated as a learned theory of logographers:157

I would then emphasize the fact that it is not till the fifth century that we hear of the
Phoenician theory, or of the connexion between Cadmus and Europa—the two car-
dinal points of the later story;—and suggest that the silence of Pindar and Aeschylus,
and perhaps of Sophocles, the insistence of Herodotus and Euripides, and the curious
variants in Pherecydes, may be significant, and mean that the theory has not long
been formulated, nor as yet universally accepted.158

Gomme readily admits that the absence of Cadmus’ Phoenician origins in all earlier
sources does not necessarily mean that this tradition did not yet exist and its absence
may be fortuitous.159 Likewise, the additional arguments adduced by Francis Vian against
a Phoenician background for Cadmus, such as the fact that he is never depicted in ‘Asiatic’
dress on vases, and that the cults he supposedly founded are Greek or pre-Greek,160 are not
conclusive either.161 By the same token, however, his ‘Phoenician’ roots cannot be proven
to be of great antiquity either. Rob Beekes adduces a number of arguments against the

152
For a well-balanced discussion, see the study of Edwards (1979), and Gruen (2010) 233–39.
153
For an overview, see Edwards (1979) 23–29; Gantz (1993) 208–10, 467–68.
154
We may further note the reference in the Suda and the lexicon of Photius, according to which Cadmus was a
son of the king of Thebes; see Edwards (1979) 49 with references.
155
Note that M. West (1985) 82–83 makes the case that Cadmus must have been mentioned already as a son of
Agenor or Phoinix in a lost section of book 3 of Hesiod’s Catalogue of Women, but this must remain hypothetical.
156
For this debate, see especially Vermeule (1971), Edwards (1979) 50–55 and the critical discussion hereof by
Beekes (2004).
157
Gomme (1913a; 1913b). Some scholars have believed in the historicity of the Phoenician element, but argued
that the term ‘Phoenician’ does not refer to Phoenicia, but may once have referred to ‘Crete’ or ‘Minoan’. The
story of Cadmus coming to Thebes was thus seen as a memory of Minoan settlement there (for discussion, see
Edwards (1979) 87–113). Though this may be attractive from a historical perspective, as remarked by Edwards,
there is no clear evidence that ‘Phoenician’ was ever used to refer to Crete. To my knowledge, it is only in a later
Egyptian source, the Canopus decree from the Ptolemaic period, that φοινίκη appears to be linked to Crete
(Altenmüller (2010) 36). Further, the fact that Cadmus is referred to as ‘the islander’ in Lucian’s Lis consonantium
5.3 might imply Cretan origins.
158
Gomme (1913a) 71–72.
159
Gomme (1913a) 67, 74.
160
Vian (1963).
161
Edwards (1979) 76–83.

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 The Journal of Hellenic Studies 247

Phoenician origins of Cadmus, such as the fact that his name (and that of his companion
Membliaros) is pre-Greek.162 He also rightly draws attention to the fact that, according to
the Suda and Photius, Cadmus is the son of a Boeotian king, Ogygus. These are admittedly
much later sources, but Ogygus is already mentioned by Pausanias.163 As aptly summarized
by Emily Vermeule, ‘it appears that there is no early connection, in the Greek mind at least,
between Kadmos and the Near East’.164
As argued by Gomme, Herodotus was prone to assuming foreign origins, and was espe-
cially fond of Phoenician middlemen.165 Instructive is the passage he quotes from Plutarch,
who accused Herodotus, among other things, of being a φιλοβάρβαρος.166 Plutarch
expresses his frustration at Herodotus’ eagerness to reject Heracles’ Greek roots:

The fact is that he has completely abandoned Epaphus and Io and Iasus and Argus; not
only is he anxious to establish an Egyptian and a Phoenician Heracles; he says that our
own Heracles was born after the other two, and he wants to remove him from Greece
and make a foreigner out of him. Yet of the learned men of old neither Homer nor
Hesiod nor Archilochus nor Peisander nor Stesichorus nor Alcman nor Pindar ever
mentioned an Egyptian or a Phoenician Heracles, but all of them know only one,
our own Heracles who is both Boeotian and Argive.167

Regrettably, Plutarch does not provide us with a similar rant about Herodotus’ treat-
ment of Cadmus, but in light of the above, it is tempting to postulate that this Greek hero
underwent a comparable process of ‘foreignization’ in the hands of Herodotus. In this case,
Herodotus may have felt forced to do so in order to reconcile the ancient tradition of
Cadmus with his own theory. After all, Cadmus, as a typical culture hero, was celebrated
for introducing phoinikeia grammata. In the original legend, these referred to Linear B writ-
ing in the Late Bronze Age. Incidentally, Linear B palm-leaf shaped tablets have been dis-
covered at Thebes, the city which, according to legend, was founded by the hero
Cadmus.168 Herodotus linked phoinikeia grammata to the Phoenician origins of the Greek
alphabet instead, which compelled him to connect Cadmus with Phoenicia. The
Phoenician roots of Cadmus are thus in all likelihood just as much the result of ‘learned
theorizing’ as the Phoenician origins of the Greek alphabet.

XIV. Unravelling the Phoenician confusion

It has been argued above that Herodotus’ explanation of the expression phoinikeia gram-
mata is based on a misinterpretation. If one adheres to the current paradigm that the
Greeks took over the alphabet from the Phoenicians, in the scenario proposed here
one would be faced with an extraordinary coincidence: the adjective used to refer to
Linear B writing (phoinikeia in the sense of ‘palm leaf’) happens to be the same as that
denoting the origin of the Greek alphabet (phoinikeia in the sense of ‘Phoenician’). It is hard
to accept such an astonishing instance of serendipity, but actually we may not have to. The

162
Beekes (2004) 177. For a different view, see West (1997) 448–50 and Astour (1967) 147, who consider the
name Cadmus to be Semitic, the latter interpreting the view of scholars who doubt his Phoenician roots as reluc-
tance to admit the Semitism of Cadmus.
163
Beekes (2004) 177. Though Pausanias does not explicitly state that he is the father of Cadmus, he does appear
to be an ancient local hero.
164
Vermeule (1971) 183; Beekes (2004) 174.
165
Gomme (1913b) 233, 236, 238.
166
Gomme (1913b) 240.
167
Plut. Mor. 857e–f, tr. Pearson (1965): 29.
168
Other accounts of the founding of Thebes circulated from early onwards; see Gantz (1993) 467.

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 248 Willemijn Waal

paradigm according to which the Greeks took over the alphabet from the Phoenicians in
the ninth or eighth century BC may be virtually undisputed within classical studies, but
this scenario has long been contested in other disciplines, notably the field of Semitic stud-
ies.169 Though the Semitic roots of the Greek alphabet are undeniable, the now available
archaeological and epigraphic data do not unequivocally point to a direct Phoenician
ancestor. There is some compelling evidence that suggests that the Greek alphabet is much
older and was transmitted from an earlier West Semitic source.170 This is hardly the place
to reopen the long-standing debate about the date of the introduction of the Greek alpha-
bet, and for now suffice it to say that one can certainly not rule out the possibility that the
Phoenicians were not the ones who introduced the alphabet to Greece, and that Herodotus
did not accidentally get it right, but rather contributed to sustaining, if not creating, an
erroneous paradigm. The present article has challenged this paradigm, proposing a differ-
ent scenario that may be summarized as follows:

• The expression phoinikeia grammata referred to Linear B, named after the pri-
mary material on which this script was written (palm leaves). The spread of this
writing system throughout the Aegean in the Late Bronze Age, which is archaeo-
logically confirmed, was connected to legendary Greek heroes like Cadmus and
Palamedes.171 Whether the term phoinikeia grammata was already in use in the
Late Bronze Age and continued to be known after this script was no longer in
use, or was coined in later times to refer to this ancient writing system, is
impossible to tell, but it was in any case in use in the fifth century BC. As is
evident from, for example, Pliny the Elder, the ancient practice of writing on
leaves was still known in the first century AD. Another term that emerged
for this ancient Aegean writing system was ‘Pelasgic letters’, referring to the
people (the ancestors of the ‘classical Greeks’) who used it.
• From at least the fifth century onwards, the original meaning of phoinikeia gram-
mata and its connection to Linear B was no longer commonly known, and a dif-
ferent theory started to circulate, which is first attested in Herodotus. The
multifunctional adjective phoinikeia was reinterpreted as ‘Phoenician’ and con-
nected to the alleged Phoenician roots of the Greek alphabet. This ‘learned
reanalysis’ was undoubtedly triggered by the hybridity of the word φοῖνιξ,
which could also mean ‘Phoenician’, and seemingly confirmed by the obvious
similarities between the Phoenician and Greek alphabets and the ubiquitous
Phoenician presence in the Aegean at that time. This theory was especially
attractive to the disposition of certain logographers such as Herodotus, who
was less interested in local traditions and more inclined to detect foreign
(Oriental) influences and who showed a great veneration for the inventiveness
of the Phoenicians. As a consequence, the expression phoinikeia grammata also
came to refer to (Archaic) alphabetic inscriptions.
• The ‘Phoenician’ theory conflated with earlier traditions featuring Cadmus,
Palamedes or other (Greek) heroes from the Heroic Age as the bringers of phoi-
nikeia grammata (i.e. Linear B), which explains their anachronistic appearance in
relation to the introduction of the alphabet. In the case of Cadmus, in all likeli-
hood it also gave rise to his Phoinikertum. Due to the reinterpretation of

169
See, for example, Ullman (1934) and Naveh (1982), but also Ruijgh (1995), Waal (2018; 2019).
170
For a recent discussion, see Waal (2018) with references.
171
It is possible that one tradition referred to the spread of Linear A, and the other of that of Linear B, but this is
impossible to reconstruct; cf. n.130 above.

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 The Journal of Hellenic Studies 249

phoinikeia grammata as ‘Phoenician letters’, referring to the alphabet, this origi-

nally Greek hero, to whom ancient legends attributed the introduction of phoi-
nikeia grammata, all of a sudden became associated with Phoenicia, which he had
not been before.
• The Phoenician reinterpretation of phoinikeia grammata generated the idea that
the alphabet reached Greece through a Phoenician intermediary, which became,
and still is, a popular theory. However, other narratives about the origins of the
Greek alphabet continued to circulate, as is evident from the fact that many
classical authors (also) provide alternative explanations.
• Though the ‘Phoenician’ interpretation was very popular, in some circles aware-
ness of the original connection between Late Bronze Age writing and phoinikeia
grammata lingered. Archaeological discoveries of ancient inscribed artefacts
may have helped to perpetuate and/or revive this knowledge. In all likelihood,
these ancient inscriptions could no longer be read, but a select group of people
(for example, in the Ismenium at Thebes) may have had a rudimentary under-
standing of its general appearance, characteristics and original usage. As some
of the discussed examples make clear, though, this was certainly not common
knowledge.
• The expression phoinikeia grammata understandably gave rise to confusion. In
some contexts, it was still used in its original meaning, referring to Late
Bronze Age writing, but this meaning was no longer understood by all. This
led to misinterpretation and the expression was taken to refer to the
(Archaic) Greek alphabet instead. To add to the bewilderment, the multifaceted
adjective φοῖνιξ was simultaneously used in its other meanings; phoinikeia gram-
mata could also refer to ‘red letters’, and to the contemporary Phoenician
alphabet.

XV. Balancing the evidence

In the above scenario, one has to assume a reinterpretation of phoinikeia grammata in or not
long before the fifth century BC. Such an assumption is generally not an attractive solu-
tion, as it has the appearance of special pleading. Considering the circumstance that vari-
ous conflicting traditions about the expression phoinikeia grammata coexisted, however, the
conclusion that at some point its meaning was falsely reinterpreted, giving rise to alter-
native narratives, is unavoidable in any scenario, including the current ‘Phoenician’ inter-
pretation. It then boils down to weighing the probabilities, which is admittedly a subjective
affair. As argued above, the proposed ‘Phoenician’ reanalysis of phoinikeia grammata around
the fifth century BC is acceptable and even likely, considering the Zeitgeist of that time. The
strong Phoenician presence, as well as the fact that the Phoenician and Greek alphabets are
obviously related, were contributing factors to its popularity. The scenario proposed here
has some undeniable advantages over the explanation of Herodotus, as it solves a number
of awkward difficulties.

• The assumption that phoinikeia grammata refers to Linear B explains why this
expression is so strongly tied to the Heroic Age, which is usually situated in
the Late Bronze Age. Moreover, the characteristics of phoinikeia grammata con-
cur exactly with those of Linear B. If one takes the expression to refer to the
Greek alphabet, these facts cannot be adequately explained and one is forced
to resort to ad hoc rationalizations.

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 250 Willemijn Waal

• The assumption that phoinikeia grammata refers to Linear B would clarify the
role of Cadmus, as well as the term ‘Pelasgic letters’. In the Phoenician inter-
pretation, by contrast, the role of Cadmus is anachronistic and the expressions
‘Pelasgic letters’ and ‘Phoenician letters’ are incompatible.
• The interpretation of phoinikeia grammata as referring to ‘palm-leaf letters’ is
also preferable from a semantic point a view: it is quite natural that terminology
for writing should be connected to (primary) writing materials (compare, for
instance, the Lontara script), whereas the use of a foreign ethnikon to refer
to this activity (if it is not contrasted to another script) would be unusual.

Ahl boldly ended his article with the following statement: ‘Cadmus was a Mycenaean,
and the writing he brought to Thebes was Linear B, which may have been known to Greek-
speaking peoples then or later as phoinikeia grammata’.172 I hope to have shown that this
scenario is indeed preferable to the Phoenician interpretation of Herodotus for a variety of
reasons. The repercussions are significant, and I would like to conclude by highlighting
three in particular. First of all, the account of Herodotus can no longer be used as evidence
for, or confirmation of, the Phoenician ancestry of the Greek alphabet. The insight that the
philological evidence adduced by Herodotus is erratic may serve as an incentive to study
the date and transmission of the alphabet to the Aegean (and beyond) without the pre-
conceived idea of a Phoenician intermediary, and to explore other scenarios which
may better account for the evidence currently available.173 Secondly, we have to accept
that, in select circles, the ancient Greeks (and Romans) had a more astute and realistic
understanding of the distant past than they are usually given credit for. Last but not least,
the fact that the Linear B script was referred to as ‘palm-leaf writing’ implies that the
written production of the Mycenaeans was much more prolific (and undoubtedly more
diverse!) than the somewhat monotonous surviving clay tablets lead one to believe.
Supplementary material. For supplementary material accompanying this paper visit https://doi.org/10.1017/
S0075426922000131

Acknowledgements. I am immensely indebted to Marco Poelwijk for his patience, support and indispensable
advice, and to Mathieu de Bakker, Luuk Huitink, Janric van Rookhuijzen, Jorrit Kelder, Alwin Kloekhorst and Bert
van der Spek for their valuable input. I would further like to thank the two anonymous reviewers and the former
JHS editor Douglas Cairns for their extremely helpful comments and suggestions. Needless to say, I alone am
responsible for the views expressed here and any errors or shortcomings that may remain.

Bibliography
Ahl, F.M. (1967) ‘Cadmus and the palm-leaf tablets’, AJPh 88.2, 188–94
Altenmüller, H. (2010) ‘Bemerkungen zum Ostfeldzug Ptolemaios’ III. nach Babylon und in die Susiana im Jahre
246/245’, in J. Fincke (ed.), Festschrift für Gernot Wilhelm anlässlich seines 65. Geburtstages am 28. Januar 2010
(Dresden) 27–44
Andolfi, I. (2019) Acusilaus of Argos’ Rhapsody in Prose: Introduction, Text, and Commentary (Trends in Classics
Supplementary 70) (Berlin and Boston)
Astour, M.C. (1967) Hellenosemitica: The Ethnic and Cultural Study in West Semitic Impact on Mycenaean Greece (Leiden)
Aura Jorro, F. and Adrados, F.R. (1993) Diccionario mycénico (Madrid)
Beckman, G.M, Bryce, T.R. and Cline, E.H. (2011) The Aḫḫiyawa Texts (Atlanta)
Beekes, R.S.P. (2004) ‘Kadmos and Europa, and the Phoenicians’, Kadmos 43.1, 167–84
—— (2010) Etymological Dictionary of Greek (Leiden)
Bennet, J. (2001) ‘Agency and bureaucracy: thoughts on the nature and extent of administration in Bronze Age
Pylos’, in S. Voutsaki and J. Killen (eds), Economy and Politics in the Mycenaean Palace States: Proceedings of a
Conference Held on 1–3 July 1999 in the Faculty of Classics (Cambridge) 25–37

172
Ahl (1967) 194.
173
Waal (2018; 2019).

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 The Journal of Hellenic Studies 251

Bernal, M. (1990) Cadmean Letters: The Transmission of the Alphabet to the Aegean and Further West before 1400 B.C.
(Winona Lake)
Bett, R. (2018) Sextus Empiricus: Against Those in the Disciplines (Oxford)
Biesantz, H. (1958) ‘Mykenische Schriftzeichen einer boötischen Schale des 5. Jahrhunderts v. Chr.’, in E. Grumach
(ed.), Minoica. Festschrift zum 80. Geburtstag van J. Sundwall (Berlin) 50–60
Bintliff, J. (2012) The Complete Archaeology of Greece: From Hunter-Gatherers to the 20th Century A.D. (Malden)
Boardman, J. (2002) The Archaeology of Nostalgia: How the Greeks Re-created Their Mythical Past (London)
Bourogiannis, G. (2019) ‘Between script and languages: inscribed intricacies from Geometric and Archaic Greek
contexts’, in P.J. Boyes and P.M Steele (eds), Understanding Relations between Scripts II: Early Alphabets (Oxford)
151–80
Bryce, T. (2003) Letters of the Great Kings of the Ancient Near East: The Royal Correspondence of the Late Bronze Age
(London and New York)
Carpenter, R. (1935) ‘Letter of Cadmus’, AJPh 56, 5–13
Ceccarelli, P. (2013) Ancient Greek Letter Writing: A Cultural History (600 BC–150 BC) (Oxford)
Chadwick, J and Baumbach, L. (1963) ‘The Mycenaean Greek vocabulary’, Glotta 41.3–4, 157–271
Chantraine, P. (1968) Dictionnaire étymologique de la langue grecque: histoire des mots (Paris)
—— (1972) ‘À propos du nom des phéniciens et des noms de la pourpre’, StudClas 14, 7–12
Collard, C. and Cropp, M. (2008a) Euripides, Fragments: Aegeus–Meleager (Loeb Classical Library 504)
(Cambridge MA)
—— (2008b) Euripides, Fragments: Oedipus–Chrysippus, Other Fragments (Loeb Classical Library 506)
(Cambridge MA)
Corcella, A. (1986) ‘Dionisio Skytobrachion, i phoinikeia e ‘l’“alfabeto pelasgico”: per una corretta interpretazione
di Diodoro III.67.1 (FGrHist 32 F 8)’, AAT 120, 41–82
Decleva Caizzi, F. (1981) Pirrone, Testimonianze (Collana di testi e studi sul pensiero antico 5) (Naples)
Devreesse, R. (1950) Codices Vaticani Graeci. Tomus III (Vatican City)
Dewald, C. (1987) ‘Narrative surface and authorial voice in Herodotus’ Histories’, in D. Boedeker (ed.), Herodotus and
the Invention of History (= Arethusa 20.1–2), 147–70
Diringer, D. (1953) The Hand-Produced Book (New York)
Dowden, K. (2016) ‘Diktys of Crete (49)’, in I. Worthington (ed.), Brill’s New Jacoby https://doi.org/10.1163/1873-
5363_bnj_a49
Driessen, J. (2000) The Scribes of the Room of the Chariot Tablets at Knossos: Interdisciplinary Approach to the Study of a
Linear B Deposit (Salamanca)
Easterling, P.E. (1985) ‘Anachronism in Greek tragedy’, JHS 105, 1–10
Edwards, G.P. and Edwards, R.B. (1974) ‘Red letters and Phoenician writing’, Kadmos 13, 48–57
—— (1977) ‘The meaning and etymology of ΠΟΙΝΙΚAΣTAΣ’, Kadmos 16, 131–40
Edwards, R.B. (1979) Kadmos the Phoenician: A Study in Greek Legends and the Mycenaean Age (Amsterdam)
Evans, A.J. (1909) Scripta Minoa: The Written Documents of Minoan Crete with Special Reference to the Archives of Knossos,
Volume 1: The Hieroglyphic and Primitive Linear Classes (Oxford)
—— (1921) The Palace of Minos: A Comparative Account of the Successive Stages of the Early Cretan Civilization as
Illustrated by the Discoveries at Knossos 1 (London)
—— (1952) Scripta Minoa: The Written Documents of Minoan Crete with Special Reference to the Archives of Knossos,
Volume 2: The Archives of Knossos. Clay Tablets Inscribed in Linear Script B Edited from Notes, and Supplemented by
John L. Myres (Oxford)
Evans, J. (1968) ‘Father of History or Father of Lies? The reputation of Herodotus’, CJ 64.1, 11–17
Fehling, D. (1971) Die Quellenangaben bei Herodot: Studien zur Erzählkunst Herodots (Berlin and Boston)
Ford, A. (1992) Homer: The Poetry of the Past (Ithaca)
Forsdyke, E.J. (1956) Greece before Homer (London)
Fowler, R.L. (2000) Early Greek Mythography I: Texts (Oxford)
—— (2013) Early Greek Mythography II: Commentary (Oxford)
Gainsford, P. (2012) ‘Diktys of Crete’, CCJ 58, 58–87
Gantz, T. (1993) Early Greek Myth: A Guide to Literary and Artistic Sources (Baltimore)
Gomme A.W. (1913a) ‘The legend of Cadmus and the logographi’, JHS 33, 53–72
—— (1913b) ‘The legend of Cadmus and the logographi-II’, JHS 33, 223–45
Gruen, E.S. (2010) Rethinking the Other in Antiquity (Princeton)
Gschnitzer, F. (2006) ‘Pelasgoi’, in H. Cancik, H. Schneider and M. Landfester (eds), Der Neue Pauly https://dx.doi.
org/10.1163/1574-9347_dnp_e912400
Guarducci, M. (1987) L’epigrafia greca dalle origini al tardo impero (Rome)
Gzella, H. (2018) review of J. Quinn (2017) In Search of the Phoenicians (Princeton), BO 75.3–4, 379–87

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 252 Willemijn Waal

Hallager, E. (1996) The Minoan Roundel and Other Sealed Documents in the Neopalatial Linear A Administration 1
(Aegaeum 14) (Liège and Austin)
Hani, J. (1980) Plutarque: œuvres morales. Tome VIII (Paris)
Hansen, W. (2003) ‘Strategies of authentication in ancient popular literature’, in S. Panayotakis, M. Zimmerman
and W.H. Keulen (eds), The Ancient Novel and Beyond (Mnemosyne Supplement 241) (Leiden and Boston) 301–14
Hawkins, J.D. (1998) ‘Tarkasnawa king of Mira: ‘Tarkondemos’. Bogazköy sealings and Karabel’, AS 41, 1–31
Hett, W.S. (1936) Aristotle, Minor Works (Loeb Classical Library 307) (Cambridge MA)
Higbie, C. (1999) ‘Craterus and the use of inscriptions in ancient scholarship’, TAPhA 129, 43–83
—— (2003) The Lindian Chronicle and the Greek Creation of Their Past (Oxford)
—— (2014) ‘Greeks and the forging of Homeric pasts’, in B. Alroth and C. Scheffer (eds), Attitudes towards the Past in
Antiquity: Creating Identities. Proceedings of an International Conference held at Stockholm University, 15–17 May 2009
(Stockholm) 9–19
—— (2017) Collectors, Scholars, and Forgers in the Ancient World (Oxford)
Hoffner, H.A. Jr. (2009) Letters from the Hittite Kingdom (Atlanta)
Hornblower, S. (2013) Herodotus, Histories: Book V (Cambridge)
Horsfall, N. (2008–2009) ‘Dictys Ephemeris and the parody of scholarship’, ICS 33–34, 41–63
Jal, P. (1995) Tite-Live: Histoire Romaine. Tome XVIII, Livre XXVIII (Paris)
Jeffery, L.H. (1967) ‘AΡΧAΙA ΓΡAMMATA: some ancient Greek views’, in W.C. Brice (ed.), Europa: Studien zur
Geschichte und Epigraphik der frühen Aegaeis (Berlin) 152–66
Jenkins, T.E. (2006) Intercepted Letters: Epistolarity and Narrative in Greek and Roman Literature (Lanham MD)
Jouan, F. and Van Looy, H. (2000) Euripide, viii.2 Fragments (Paris)
Kannicht, R. (2004) Tragicorum graecorum fragmenta, Vol. 5: Euripides (Göttingen)
Krzyszkowska, O. (2005) Aegean Seals: An Introduction (London)
Lachenaud, G. and Coudry, M. (2011) Cassius Dio, Histoire romaine. Livres 38, 39 & 40 (Paris)
Larson, J. (1995) Greek Heroine Cults (Madison)
Lehmann, R.G. (2019) ‘Much ado about an implement!—the Phoenicianising of early alphabetic’, in P.J. Boyes and
P.M Steele (eds), Understanding Relations between Scripts II: Early Alphabets (Oxford) 69–90
Mahadevan, I. (2003) Early Tamil Epigraphy: From the Earliest Times to the Sixth Century A.D. (Cambridge MA)
Marinatos, S. (1958) ‘Γραμμάτων διδασκάλια᾽, in E. Grumach (ed.), Minoica: Festschrift zum 80. Geburtstag van J.
Sundwall (Berlin) 226–31
Melchert, H. C. (forthcoming) ‘Mycenaean and Hittite diplomatic correspondence: fact and fiction’, in
A. Teffeteller (ed.), Proceedings of the Workshop Mycenaeans and Anatolians in the Late Bronze Age Held in
Montreal, Quebec, January 4–5, 2006
Melena, J.L. (1975) ‘po-ni-ki-jo in the Knossos Ga tablets’, Minos 14, 77–84
Merkle, S. (1989) Die Ephemeris belli Troiani des Diktys von Kreta (Studien zur klassischen Philologie 44) (Frankfurt am
Main and New York)
Most, G.W. (2018) Hesiod, Volume II: The Shield, Catalogue of Women, Other Fragments (Loeb Classical Library 503)
(Cambridge MA)
Mylonas, G.E. (1966) Mycenae and the Mycenaean Age (Princeton)
Nadaff, G. (2003) ‘Anthropogony and politogony in Anaximander of Miletus’, in D.L. Couprie, R. Hahn and G. Naddaf
(eds), Anaximander in Context: New Studies in the Origins of Greek Philosophy (SUNY Series in Ancient Greek
Philosophy) (Albany) 9–71
Naveh, J. (1982) Early History of the Alphabet: An Introduction to West Semitic Epigraphy and Palaeography (Jerusalem
and Leiden)
Ní Mheallaigh, K. (2008) ‘Pseudo-Documentarism and the limits of ancient fiction’, AJPh 129.3, 403–31
—— (2012) ‘The “Phoenician Letters” of Dictys of Crete and Dionysius Scytobrachion’, The CCJ 58,
181–93
Oldfather, C.H. (1953) Diodorus Siculus, Library of History, Volume II: Books 2.35–4.58 (Loeb Classical Library 303)
(Cambridge MA)
—— (1956) Diodorus Siculus, Library of History, Volume IV: Books 9–12.40 (Loeb Classical Library 375) (Cambridge MA)
Owen, D.I and Young, G.D (1997) ‘An interview with Michael Astour on the occasion of his being honored by the
Middle West Branch of the American Oriental Society, February 11 and 12, 1996’, in G.D. Young, M.W. Chavalas
and R.E. Averbeck (eds), Crossing Boundaries and Linking Horizons: Studies in Honor of Michael C. Astour (Bethesda)
1–36
Padmakumar, P.K., Sreekumar, V.B., Rangan, V.V. and Renuka, C. (2003) ‘Palm leaf writing’, Kerala 47.3, 125–29
Palaima, T. (2003) ‘“Archives” and “scribes” and information hierarchy in Mycenaean Greek Linear B records’, in
M. Brosius (ed.), Ancient Archives and Archival Traditions: Concepts of Record-keeping in the Ancient World (Oxford
and New York), 153–94
Palmer, L.R. (1957) ‘A Mycenaean tomb inventory’, Minos 5, 58–92

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 The Journal of Hellenic Studies 253

Papazarkadas, N. (2014) ‘Two new epigrams from Thebes’, in N. Papazarkadas (ed.), The Epigraphy and History of
Boeotia: New Finds, New Prospects (Leiden and Boston) 223–51
Parker, R. (2010) ‘Aegesilaus and the bones of Alcmena’, in H.-G. Nesselrath, D. Russell, G. Cawkell, W. Deuse,
J. Dillon, R. Parker, C. Pelling and S. Schröder (eds), On the Daimonion of Socrates: Plutarch (Tübingen) 129–37
Perna, M. (2011) ‘Minoan and Mycenaean archives: the problem of documents on perishable material’, MediChron
1, 9–25
—— (2017) ‘Administrative documents without writing: the case of sealings and flat-based nodules’, in
S. Ferrara, A.M. Jasink and J. Weingarten (eds), Non-scribal Communication Media in the Bronze Age Aegean and
Surrounding Areas (Florence) 73–80
Pearson, L. (1965) Plutarch: Moralia, Volume XI: On the Malice of Herodotus, Causes of Natural Phenomena (Loeb Classical
Library 426) (Cambridge MA)
Phillips, E.D. (1957) ‘A suggestion about Palamedes’, AJPh 78, 267–87
Quinn, J. (2017) In Search of the Phoenicians (Princeton)
Roth, C. (2002) The Suda On Line: Byzantine Lexicography http://www.stoa.org/sol/
Ruijgh, C. (1967) Études sur la grammaire et le vocabulaire du grec mycénien (Amsterdam)
—— (1995) ‘D’Homère aux origines proto-mycéniennes de la tradition épique. Analyse dialectologique du lange
homérique, avec un excursus sur la création de l’alphabet grec’, in J.P. Crielaard (ed.), Homeric Questions: Essays
on Philology, Ancient History and Archaeology, Including Papers of a Conference Organized by the Netherlands Institute at
Athens (15 May 1993) (Amsterdam) 1–96
Schachter, A. (1981) Cults of Boiotia, 1: Acheloos to Hera (Bulletin Supplement 38.1) (London)
Scheer, T.S. (1996) ‘Ein Museum griechischer “Frühgeschichte” im Apollontempel von Sikyon’, KLIO 78.2, 353–73
Schmidt, N. (1920) ‘Bellerophon’s tablet and the Homeric question in the light of Oriental research’, TAPhA 51,
56–70
Schneider, J. (2004) ‘Les traditions relatives à l’origine de l’alphabet grec dans les scholies de la TEΧΝΗ de Denys le
Thrace’, Bulletin de l’Association Guillaume Budé Année, 120–57
Schwartz, E. (1887) Scholia in Euripidem 1 (Berlin)
Schwartz, J. (1958) ‘Le tombeau d’Alcmène’, RA 1, 76–83
Schnapp, A. (1996) The Discovery of the Past (London)
Scodel, R. (1980) The Trojan Trilogy of Euripides (Hypomnemata 60) (Göttingen)
Shear, I.M. (1998) ‘Bellerophon tablets from the Mycenaean world? A tale of seven bronze hinges’, JHS 118, 187–89
Steele, P. M. (2017) ‘Writing “systems”: literacy and the transmission of writing in non-administrative contexts’, in
A.M. Jasink, J. Weingarten and S. Ferrara (eds), Non-scribal Communication in the Bronze Age Aegean and
Surrounding Areas: The Semantics of A-literate and Proto-literate Media (Florence) 153–72
Steiner, D. (1994) The Tyrant’s Writ: Myth and Images of Writing in Ancient Greece (Princeton)
Stott, K.M. (2008) Why Did they Write this Way? Reflections on References to Written Documents in the Hebrew Bible and
Ancient Literature (New York)
Strauss-Clay, J. (2016) ‘Homer’s epigraph: Iliad 7.87–91’, Philologus 160.2, 185–96
Surenhagen, D. (2008) ‘Schriftsysteme und Schriftgebrauch in der Ägäis, dem östlichen Mittelmeerraum und im
Alten Orient während des 2. Jahrtausends v. Chr.’, in A. Kyriatsoulis (ed.), Austausch von Gütern, Ideen
und Technologien in der Agäis und im östlichen Mittelmeer. Von der prähistorischen bis zur archaischen
Zeit: [5. Internationale] Tagung, 19.–21.05 2006 in Ohlstadt Obb. Deutschland (Weilheim), 245–282
Symeonoglou, S. (1985) The Topography of Thebes from the Bronze Age to Modern Times (Princeton)
Thomas, R. (2000) Herodotus in Context: Ethnography, Science and the Art of Persuasion (Cambridge)
Thonemann, P. (2016) ‘Croesus and the oracles’, JHS 136, 152–67
Ullman, B.L. (1934) ‘How old is the Greek alphabet?’, AJA 38.3, 359–81
van den Broek, R. (1972) The Myth of the Phoenix According to Classical and Early Christian Traditions (Leiden)
van Rookhuizen, J.Z. (2019) Herodotus and the Topography of Xerxes’ Invasion: Place and Memory in Greece and Anatolia
(Berlin and Boston)
Ventris, M. and Chadwick, J. (1973) Documents in Mycenaean Greek with a Foreword by Alan J.B. Wace. 2nd edition by
John Chadwick (Cambridge)
Vermeule, E. (1971) ‘Kadmos and the dragon’, in D.G. Mitten, J.G. Pedley and J.A. Scott (eds), Studies Presented to
George M.A. Hanfmann (Mainz) 177–88
Vian, F. 1963. Les origines de Thèbes. Cadmos et les Spartes (Études et Commentaires XLVIII) (Paris)
Vignolo Munson, R. (2005) Black Doves Speak: Herodotus and the Languages of Barbarians (Hellenic Studies 9)
(Cambridge MA and London)
Waal, W. (2018) ‘On the ‘Phoenician letters’: the case for an early transmission of the Greek alphabet from an
archaeological, epigraphic, and linguistic perspective’, Aegean Studies 1, 83–125

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press

 254 Willemijn Waal

—— (2019) ‘Mother or sister? Rethinking the origins of the Greek alphabet and its relation to the other “western”
alphabets’, in P.J. Boyes and P.M Steele (eds), Understanding Relations between Scripts II: Early Alphabets (Oxford)
109–24
—— (2021) ‘Distorted reflections? Writing in the Late Bronze Aegean in the mirror of Anatolia’, in M. Bianconi
(ed.), Linguistic and Cultural Interactions between Greece and the Ancient Near East: In Search of the Golden Fleece
(Leiden and Boston) 197–232
Weingarten, J. (1983) ‘The use of the Zakro sealings’, Kadmos 22, 7–13
West, M.L. (1997) The East Face of Helicon: West Asiatic Elements in Greek Poetry and Myth (Oxford)
—— (1985) The Hesiodic Catalogue of Women: Its Nature, Structure, and Origins (Oxford)
West, S. (1985) ‘Herodotus’ epigraphical interests’, CQ 35.2, 278–305
Willi, A. (2005) ‘Kadmos anethke. Zur Vermittlung der Alphabetschrift nach Griechenland’, MH 62.3, 162–71

Cite this article: Waal, Willemijn (2022). Deconstructing the Phoenician myth: ‘Cadmus and the palm-leaf tablets’
revisited. The Journal of Hellenic Studies 142, 219–254. https://doi.org/10.1017/S0075426922000131

https://doi.org/10.1017/S0075426922000131 Published online by Cambridge University Press