At Phaedrus 238b7–c4 Plato offers a playful etymological explanation of erōs. After an extended periphrasis emphasizing its overpowering and dominating nature (kratēsasa, rhōstheisa, nikēsasa, 238b8–c3), he states that it “receives its name from that very strength (rhōmē, 238c3)” and is therefore called erōs (238c4). Compared to the highly elaborated protasis of this sentence, the apodosis at 238c4 is characterized by conspicuous brevity (Denniston 1965, 68), stylistically underscoring the overmastering force of erōs. The derivation of erōs from rhōmē is reinforced by a striking figura etymologica, the cognate cluster errōmenōs rhōstheisa (“strongly strengthened”) at 238c2. Some commentators accordingly speak of a shared ‘rhō- stem’ (e.g., Robin 1933, 19n.1; Cobb 1993, 196; Nichols 1998, 40n.1), though the derivation is paronomastic rather than morphological, exploiting the phonetic proximity of erōs with rhōmē, errōmenōs, and rhōstheisa. This association through the initial rhō- sound is sufficient to invite Plato’s wordplay.

Ancient and modern readers have long found this etymology problematic. It has been described as “strange” (Heindorf 1802, 224), “somewhat fanciful” (Fowler 1914, 447n.1), or even “most fanciful” (Brown & Coulter 1979, 244n.8), “fantasy” (Robin 1933, 19n.1), “absurd” (Shorey 1933, 199), “arbitrary” (Martini 1935, 31n.1), “strained” (Brown & Coulter 1979, 240), “obviously unserious” (Rowe 1986, 157), “farfetched and imaginative” (Nehamas & Woodruff 1995, 18n.41), and “highly suspect” (Rhodes 2003, 452).

Despite this broad agreement, scholarly opinion diverges as to what, precisely, Plato takes erōs to be derived from. Some commentators trace the name to errōmenōs rhōstheisa alone (e.g., Heindorf 1802, 224; Ast 1829, 316; Stallbaum 1857, 54). Yunis (2011, 114) likewise suggests that the primary cue is errōmenōs rhōstheisa, with the appeal to rhōmē functioning as a recapitulary echo. Others connect erōs with rhōmē alone (e.g., Hamilton 1973, 37n.1; Brown & Coulter 1979, 240; 243; 244n.8; Burger 1980, 135; Nehamas & Woodruff 1995, 18n.41; Rhodes 2003, 452; Scully 2003, 17n.47), while most take Plato’s etymology to involve both rhōmē and the cognate phrase errōmenōs rhōstheisa together (e.g., Fowler 1914, 447n.1; Robin 1933, 19n.1; Martini 1935, 31n.1; Hackforth 1952, 39n.3; Cobb 1993, 196; Heitsch 1997, 21n.9; Sansone 2007, 755; Hartmann 2017, 48n.146; 249–50; 267; 324; Emlyn-Jones & Preddy 2022, 385n.38). Among ancient commentators, the situation is similar: Proclus (in Crat. 16, 7.6–9 Pasquali; in Parm. 19.9–13 Steel) derives erōs from rhōmē alone, whereas Hermias of Alexandria (in Phdr. 3.12–14; 53.14–22; 53.24–6 Couvreur = 3.23–5; 57.7–16; 57.18–20 Lucarini/Moreschini) additionally invokes errōmenōs rhōstheisa. In any case, this distinction is largely negligible, since rhōmē, errōmenōs, and rhōstheisa form a close cognate cluster within the rhō- word-family, and Plato’s etymological play depends primarily on this shared sound pattern.

Plato’s use of the adverb errōmenōs in the definition of erōs at Phaedrus 238b7–c4 does more than intensify the lover’s compulsive desire: it already anticipates the dialogue’s later concern with rhetoric (cf. Irani 2017, 118–9). The term suggests not only the lover’s forceful and domineering disposition toward the beloved, but also a distinctive ethos of overpowering persuasion. Since the only other occurrence of errōmenōs in the Phaedrus appears in the second half, where rhetoric is described as exerting a “very forceful” (mala errōmenēn, 268a3) power over crowds, Plato subtly links erotic compulsion with rhetorical domination from an early stage in the dialogue. 

Moreover, the account presented in the Phaedrus’ first speech—where erōs is treated as a form of appetitive desire (epithumia) oriented toward bodily pleasure—represents a deliberately limited and ultimately inadequate conception within the dialogue (cf. Irani 2017, 133–4). Socrates later regrets this reduction of erōs to pleasure-seeking appetite, and the palinode corrects it by reinterpreting erōs as something divine rather than merely compulsive or self-interested. The passage at 238b7–c4 thus functions as a provisional definition that Socrates will explicitly abandon, making it clear that Plato’s fuller account of erōs cannot be captured in purely appetitive terms. This definition of erōs in terms of rhōmē echoes Plato’s rival Isocrates (Helen 55, 268.22–3 Mandilaras: τῶν καλῶν ἔρως … ῥώμην ἔχων, “erōs of the beautiful ones … possessing strength”; cf. Brown & Coulter 1979, 244n.8; Chitchaline 1992, 227; Hartmann 2017, 324). Thus, the passage is best read as a pointed parody of Isocrates’ rhetoric.

In the palinode, at Phaedrus 252b8–9, Plato will offer an alternative etymology of erōs, while in the Cratylus, he derives erōs in yet another way. These etymological accounts of erōs differ markedly both linguistically and conceptually. At Phaedrus 238b7–c4, erōs is first explained paronomastically from rhōmē (“strength”), reinforced by errōmenōs rhōstheisa (“strongly strengthened”). At 252b8–9, erōs is then reimagined through poetic wordplay as Pterōs (“Winged One”), a divine name formed by consonantal grafting and embedded in the mythic opposition of mortal and immortal language. In the Cratylus (420a9–b4), finally, erōs is derived quasi-philologically from eisrhein (“to flow in”), supported by an archaic form (esros) and a putative sound change. Conceptually, these three derivations foreground erōs respectively as an overpowering appetitive force, a winged divine power with erotic overtones, and an external influx into the lover’s soul.