1. The Problem

Homeric Greek nēpios, at first glance, does not present obvious semantic problems. LSJ [1] give as a first meaning “infant, child,” citing
νήπιον, οὔ πω εἰδόθ’ ὁμοιΐου πολέμοιο [2]

nēpios, who knew nothing yet of the joining of battle [3]
IX 440 [4]
and the frequently used phrase nēpia tekna (“nēpios children,” ΙI 136, etc.). Chantraine concurs, giving as a first meaning of nēpios “tout jeune.” [5] Various Homeric passage may be cited in which nēpios seems to mean simply “child,” for instance, where Telemachos contrasts his past and present states:
…ἐγὼ δ᾽ ἔτι νήπιος ἦα;
νῦν δ᾽ ὅτε δὴ μέγας εἰμὶ καὶ ἄλλων μῦθον ἀκούων
πυνθάνομαι, καὶ δή μοι ἀέξεται ἔνδοθι θυμός...

...when I was still nēpios?
But now, when I am grown big, and by listening to others
can learn the truth, and the anger rises within me…
ii 313–315
Metaphorically, according to LSJ, the basic meaning “child” is applied to “the understanding,” and nēpios comes to mean “childish, silly.” The following Homeric passages are cited:
νήπιóς εἰς, ὦ ξεῖν᾽, ἢ τηλόθεν εἰλήλουθας...

You are nēpios, O stranger, or else you have come from far away…
[not to know that you are on the island of Ithake] {1|2}
xiii 237
ὧς φάτο λισσόμενος μέγα νήπιος...

So he spoke, supplicating, greatly nēpios
[for he did not know he was asking for his own death]
XVI 46
... τοὶ δὲ–μέγα νήπιοι!–οὐκ ἐπίθοντο.

… but they were greatly nēpios and would not listen …
[of Odysseus’s men who would not leave the land of the Kikones]
ix 44
LSJ add “without foresight, blind,” citing XXII 445. Here Andromache is called nēpios because she ordered a bath to be prepared for Hektor, not knowing that he was already dead. Chantraine adds to his basic definition (“tout jeune”) “sot, sans raison.” [6] Frisk simply lists a variety of possible translations of nēpios, but by placing the meaning “child” at the beginning of his list, and those meaning “foolish” toward the end, he too implies that the basic meaning is “child” and that other meanings are derivative. [7]
The word child in modern English refers to either age or descent. Greek pais (“child”) has a similar semantic range. Teukros, having shot his bow, retreats behind the shield of Ajax pais hōs hupo mētera (“like a child behind its mother,” VIII 271). On the other hand, Achilles is Thetidos pais (“Thetis’s child,” IV 512, etc.), Hektor is Priamoio pais (“Priam’s child,” III 314, etc.), and so forth. The words teknon and tekos mean “child” in the sense of “offspring.” Athene, for example, is called Dios tekos, (“Zeus’s child,” II 157, etc.).
Both pais and teknon may be used figuratively. In the first of the two following examples, Nestor is encouraging the sentinels guarding the Greek camp; in the second, the herald Idaios stops the fight between Aias and Hektor:
οὕτω νῦν, φίλα τέκνα, φυλάσσετε ...

Continue to keep your watch this way, beloved children …
X 192 {2|3}
μηκέτι, παῖδε φίλω, πολεμίζετε μηδὲ μάχεσθον·

Stop the fight, dear children, nor go on with this battle.
VII 279
By addressing adults with the term “children,” the speaker is expressing his own “fatherly” concern for them.
The verbal derivative of pais, paizō, is used of Nausicaa and her companions playing or dancing on the beach (vi 106, vii 291). It also describes the activity within Odysseus’s halls at xxiii 147 (where Lattimore translates it “celebrating”). In the Homeric Hymns, also, paizō means “dancing” or similar, but perhaps unstructured, playing. In later Greek, it may mean skilled dancing or the “playing” of an instrument. In none of these figurative uses do we see disapproval or censure, which is clearly implied by the word nēpios in, for example:
αὐτῶν γὰρ σφετέρησιν ἀτασθαλίῃσιν ὄλοντο,
νήπιοι, οἳ κατὰ βοῦς Ὑπερίονος Ἠελίοι

… they were destroyed by their own recklessness,
nēpioi, who devoured the oxen of Helios, the Sun God …
i 7–9
Thus, if nēpios means “child,” it means “child” in some other sense than pais or teknon.
In English, the adjective childlike is used in a positive sense, while childish is not. It is not, in itself, surprising that words meaning “child” should have both positive and negative associations in Homeric Greek. In fact, despite the connotations of paizō, the most usual characterization of children in Homer emphasizes their weakness. [8] Diomedes taunts Paris, who has wounded him, with the following words:
οὐκ ἀλέγω, ὡς εἴ με γυνὴ βάλοι ἢ πάϊς ἄφρων·

I care no more than if a witless child or a woman had struck me …
XI 389 {3|4}
And Odysseus says of the Greeks:
ὥς τε γὰρ ἢ παῖδες νεαροὶ χῆραί τε γυναῖκες
ἀλλήλοισιν ὀδύρονται οἶκόνδε νέεσθαι

For as if they were young children or widowed women
they cry out and complain to each other about going homeward.
II 289–290
In both these passages, children are grouped with women to form a class of those who are ineffectual. [9] To compare a warrior to a woman is an insult. For example, Hektor taunts Diomedes:
Τυδεΐδη, περὶ μέν σε τίον Δαναοὶ ταχύπωλοι
ἕδρῃ τε κρέασίν τε ἰδὲ πλείοις δεπάεσσι·
νῦν δέ σ᾽ ἀτιμήσουσι· γυναικὸς ἄρ᾽ ἀντὶ τέτυξο.

Son of Tydeus, beyond others the fast-mounted Danaans honoured you
with pride of place, the choice meats and the filled wine cups.
But now they will disgrace you, who are no better than a woman.
VIII 161–163
The mealtime honors the Greeks bestow on Diomedes are precisely those that distinguish a warrior (cf. XII 310–321), and he is distinguished, especially from the women and children. When Hektor calls Diomedes a woman, he is calling him “not a warrior.” Similarly, in the passages cited above (XI 389 and II 289–290), men are called “like women and children” because they are behaving in an unwarriorlike fashion.
The epithet nēpios, however, does not present a contrast to warriorlike behavior. When Patroklos begged Achilles to allow him to lead the Myrmidons into battle, he made a bad mistake:
ὣς φάτο λισσόμενος μέγα νήπιος· ἦ γὰρ ἔμελλεν
οἷ αὐτῷ θάνατόν τε κακὸν καὶ κῆρα λιτέσθαι.

So he spoke supplicating, greatly nēpios: this was
his own death and evil destruction he was entreating.
XVI 46–47
But certainly he was not behaving in a way that would place him in the class of women and children as opposed to the class of warriors. {4|5} Nor is his mistake due to any apparent childish qualities such as weakness of reason or carelessness of signs.
Another context of nēpios, suggests that the word, in itself may not mean “child” at all. When the steward of Menelaos hesitates to offer hospitality to Telemachos and Peisistratos, Menelaos rebukes him thus:
οὐ μὲν νήπιος ἦσθα, Βοηθοΐδη ᾽Ετεωνεῦ,
τὸ πρίν᾽‧ ἀτὰρ μὲν νῦν γε πάϊς ὣς νήπια βάζεις.
Eteoneus, son of Boethoos, you were never
nēpios before, but now you are babbling nēpios things,
[or “speaking nēpios words”] as a child would do.
iv 31–32
The semantic history of Latin infans provides a model for a semantic development of nēpios alternative to LSJ’s extension of its {5|6} meaning by metaphor. [10] Latin infans is etymologically transparent, being the negative of the present participle of for (“I speak”). Cicero uses it with its simple root meaning:
scribit Herodotus, Croesi filium, cum infans esset, locutum

Herodotus writes that Croesus’s son, although he had been mute, spoke …
Cicero Div. 1.53.121
It was more commonly used, in classical Latin, as a descriptor of young children, of whom inability to speak is one distinguishing characteristic. Eventually, this characteristic epithet became a noun signifying young children, and reference to speaking ability dropped out, semantically speaking. The French word enfant implies nothing about verbal ability, and, in English, a person is infant before the law up until the age of majority. Such substantivization of adjectives, with ellipsis of a traditionally associated noun, occurs in Greek, also: hoi thnētoi (“mortals’’) < hoi thnētoi anthrōpoi (“mortal humans”), to xeinēion (“the hospitality [gift]”) < to xeinēion dōron (“the hospitality gift”), hugrē (“sea”) < hugra keleutha (“watery ways”).
In Modem Greek, n ēpion means “child” in an unmarked sense—like enfant in French (a nēpiagōgeion is a “kindergarten”). [11] In Homeric Greek, it is characteristic of children to be nēpios, just as it is characteristic of ships to be black, Achaians to be well-greaved, and so forth. [12] But the traditional association of the word nēpios with children does not tell us anything about its meaning, any more than we could deduce the particular meaning of thnētoi from all that we know of human beings. In fact, if nēpios, originally meaning something about the intellect, came to mean simply “child,” then its semantic development is curiously paralleled in the modern Greek {6|7} word mōro (“child”) which is clearly derived from mōros (“silly,” “stupid”).
The semantics of nēpios are all the more open to question since the word has no generally accepted etymology. [13] There have been various proposals, most of which can be discounted because they rely on the assumption that there existed, in very early Greek, a negative prefix -, which is not the case. [14] The only real possibilities seem to be: (1) a connection with the root wek w- found in Greek epos (“word, speech”), which would make nēpios semantically equivalent to Latin infans; [15] or (2) a derivation from the root āp- found in Latin apīscor (“reach,” “attain”) and Sanskrit āpnóti (“reach,” “attain”). [16]
Neither of these suggestions is without problems. In the case of a connection with epos, there is simply no evidence that the root originally began with the laryngeal necessary for the formation of -. More important, despite what seems like the attractive parallel with infans; there is no good semantic evidence to support such a derivation. [17] Facility in speaking, as an adult characteristic, sometimes appears in Homer in contrast to nēpios, for example:
νήπιος οὔτε πόνων εὖ εἰδὼς οὔτ᾽ ἀγοράων

nēpios all unversed in fighting and speaking
iv 818
Still there is nothing to suggest that inability to speak is the essential characteristic of a person who is nēpios. Far more common, in fact, {7|8} is the suggestion that a nēpios person is unable to know or understand the plans of the gods. For example:
νήπιος, οὐδὲ τὰ ᾔδη ἅ ῥα Ζεὺς μήδετο ἔργα

nēpios, who knew nothing of all the things Zeus planned
to accomplish …
II 38
The second possibility, that nēpios derives from the root āp- found in Latin apiscor and Sanskrit āpnóti, is also not problem-free. It seems likely that another cognate of these words can be found in the Hittite epmi (“to take”), and if this is the case, the original laryngeal in the root would be ǝ 1, (hence * ǝ 1 ep- / * ǝ 1 p-), which would give rise to an e- vocalism, thus: * 1 p- ios > nēp-ios. [18] On the other hand, the Mycenaean name na-pu-ti-jo, which some interpret as the formal equivalent of Homeric (etc.) nēputios, and the word napia, which may occur in a letter of Epicurus, would suggest an a- vocalism: thus * 2 p- ios > *nāp- ios > (Ion. etc.) nēpios. There are, however, uncertainties surrounding the validity of both these forms. In the Mycenaean syllabary, there is no positive indication as to what consonant, if any, a syllable ends with, nor whether the

of a form like <na-pu-ti-jo> is unvoiced [p], voiced [b] or aspirated [ph]. The readings of individual words, therefore, must be determined by a combination of phonetic possibilities and contextual indications. In the case of a name, contextual indications are, of course, lacking. In short, the proper reading of this spelling may or may not be Naputios. The Epicurean napia, while it appears to be what is written on the papyrus fragment, may represent a hyperdorism, and even the reading napia, is not generally accepted. [19]

Although there is no definitive reason to reject the derivation of nēpios from the root *āp- it would be easier to accept if we could find another derivative of this root in Greek with some semantic {8|9} relationship to nēpios. One candidate is haptō (“to fasten,” “join”), but there is general agreement that the spiritus asper renders this an impossibility. [20] The only other evident possibility is ēpios (“gentle,” “kind”), the derivation of which from this root is not likely to be rejected on phonological grounds. [21] Its semantic connections with apiscor and, indeed, with nēpios would have to be made clear. The following chapters suggest the possibility of a semantic connection, in Homeric diction, between ēpios and nēpios. {9|10}


[ back ] 1. LSJ = Liddell, Scott, Jones, et al., A Greek-English Lexicon9 (Oxford 1940) and Supplement (1968).
[ back ] 2. I have used Monro, D.B. and Allen, T.W., eds. Homeri Opera (lliadis) 3rd ed. (Oxford 1920) and Stanford, W.B., ed. The Odyssey of Homer 2nd ed. (London 1959).
[ back ] 3. In general, I use Lattimore’s translations (see bibliography), reserving the right to make changes where it seems necessary for emphasis. The most common change is that key words, such as nēpios are not translated.
[ back ] 4. Books of the Iliad are designated by upper-case Roman numerals and those of the Odyssey by lower-case Roman numerals.
[ back ] 5. Chantraine 1974, s.v. nēpios.
[ back ] 6. ibid.
[ back ] 7. Frisk 1970. s.v. nēpios. Frisk’s list runs as follows: “noch unmündig, jung, schwach, kindisch, unverständig, töricht.”
[ back ] 8. For a useful summary of Homeric evidence concerning childhood, see Denoël 1928.
[ back ] 9. For more on women and children as a class distinct from the class of warriors, see the first section of Chapter Three.
[ back ] 10. This is not to say that nēpios is semantically equivalent to infans.
[ back ] 11. Working backward in time through the Greek language: in the New Testament, nēpios usually means “child,” “infant” and is a substantive. The word is not very common in the prose of the classical period. Aristotle uses it substantively to mean “infant,” as does Plato, except where he is quoting a proverb (e.g. Symp. 222b7). Similarly in Herodotus, where nēpios is both an adjective and a substantive, it means simply “infant” except in quotations of oracles, where it has the sense LSJ calls “metaphorical.” In Euripides, nēpios usually means “child.” In Sophocles and Aeschylus, however, and in Pindar and the Presocratics, the so-called metaphorical sense predominates.
[ back ] 12. Of 76 occurrences of nēpios in the Iliad and Odyssey, once it modifies “words” (not expressed), once biē (“strength”), 35 times it is used of adults, and 39 times of children, of which the phrase nēpia tekna accounts for 14.
[ back ] 13. The principal suggestions are °Osthoff and Brugman 1881 derive nēpios from a root pu- (cf. Sanskrit pu- nā- ti “reinigen,” “aufklaren”). Thus *nē- pw- io- s > nēpios. °Specht 1928: nēpios < + *ap- (cf. *apel-, root meaning “Kraft, Stärke”). °Rozwadowski, summarized in Lacroix 1937.261, connects nēpios and ēpios with Latin pius. Thus ēpios < ē + *pwiios “avec le meme ē- initial que dans ēbaios à côté de baios. °Lacroix 1937 also connects ēpios and nēpios, i.e. nēpios < *ne- ēpios, the root of which the author wants to connect with ēpuō. But we have Doric āpuo and not Doric *āpios. A stronger objection in Beekes 1969.106: “There is … no trace of Greek *ne- from PIE *ne-.
[ back ] 14. Beekes 1969.98–113.
[ back ] 15. Puhvel 1953.25.
[ back ] 16. Pisani 1939: *ne- apios > nēpios (but see final comment in note 12).
[ back ] 17. cf. Chantraine 1974 s.v. nēpios: “le sens ‘qui ne sait pas parler’ ne s’appuie sur aucune tradition ancienne.”
[ back ] 18. Beekes’s reservation about the derivation from the root *ap- is that there seems to be reasonable evidence for both ǝ1 and ǝ2.
[ back ] 19. The papyrus reading is ΝΑΠΙΑ. Usener (Teubner 1887) and Baily (OCT 1926) read ἡ αἰτία. The context is a letter to a child; the text runs: εὖ γὰρ ἴσθι, ἡ αἰτία [or νάπια] ὅτι καὶ ἐγω καὶ ο[ἱ] λοιποὶ πάντες σε μέγα φιλοῦμεν, ὅτι τούτοις πείθῃ πάντα … The disputed reading might also be a proper name.
[ back ] 20. See, e.g., Meillet 1929.276.
[ back ] 21. See Lacroix 1937.261. But the formation of the negative nēpios must be far older than he imagines—i.e. both nēpios and ēpios must have existed independently since before the disappearance of the laryngeal.