Interchange of f- and h-

 

Formiae = Hormiae = Formia in Southern Latium

I thought a few more words could be said about the interchange of initial f- and h- which I cover on p. 505 of the 2nd edition. These paragraphs replace what I wrote there. 

G. There are a number of cases where we find initial f for expected h < *ĝh/*gh and also a few cases where we find initial h for expected f from *dh and *bh. (67) First, the case of h for f: 

horda ‘pregnant cow’ Var. R. 2.5.6 whence the festival Hordicidia (Paul. Fest. p. 91 L) = forda (Var. L. 6.15, etc.) from the root *bher- ‘bear’.

 

hanula parua delubra, quasi fanula “A hanula is a small shrine, as it were, a fanula” (Paul. Fest. p. 91 L) < *dhh1sno- ‘shrine’. 

hebris ‘fever’ (Serv. ad Aen. 7.695) = febris <*dhegwhris, and attributed to the antiqui. 

horctum et forctum pro bono dicebant (Paul. Fest. p. 91 L; CGL 5.503,35 and 570,17). horctus and forctus are archaic o-stem variants for the standard i-stem form fortis ‘strong’. (68)

haba ‘bean’ = faba < *bhabheh2, cf. OCS bobŭ ‘bean’. Faliscan or used by the antiqui. (69) 

and then the cases of f for original h:

fircus ‘he-goat’ = hircus (“Sabine” or antiqui). 

fēdus ‘goat’ = haedus (70) (“Sabine” or antiqui), cf. OE gāt ‘goat’. 

fasēna ‘sand’ = harēna (71) (“Sabine”). 

The h for f cases are either unattributed or attributed to the antiqui. The f for h cases are attributed to “Sabine” to the antiqui or to Faliscan and in fact in Faliscan initial f- became h- in the Middle Faliscan period: 

hileo ‘son’ (Bakkum 146) = fīlius. (72)   

and we also find hypercorrect f in Faliscan and in Praenestine: 

foied ‘today’ (Bakkum 59) = hodiē. 

fe ‘here’ (Bakkum 56) = hīc. (73)  

FERCLES = Herc(u)lēs (CIL 1.2.564, Praeneste). (74)

We may infer that there were a number of dialects of Latin, including at least Faliscan, that weakened f to h in initial position and that a hypercorrective reaction led to the creation of forms with f for original h. (75)  This confusion was well known and some examples are quite possibly artificial creations in the service of folk-etymological theorizing. (76) 

 _______________________________ 

(67) See Hiersche 1965 for a collection and sifting of examples. 

(68) Elsewhere Paul. Fest. p. 74 L. defines forctes as frugi et bonus siue ualidus and in a discussion of a tribal name of old Latium Forctes Festus p. 474 L comments fortibus id est bonis. This adjective is of uncertain etymology but the form with f is certainly old and has a cognate or was borrowed into Osc. fortis ‘stronger’ (TB 12).

(69) Attributed to Faliscan by Terentius Scaurus (Keil 7.130): utraque (f and h) enim ‹est› flatus; quare quem antiqui fircum nos hircum et quam Falisci habam nos fabam appellamus, et quem antiqui fariolum nos hariolum. “For both (f and h) are aspirated; So what the ancients called fircus we call hircus, and what the Faliscans call haba we call faba, and what the ancients called fariolus we call hariolus.”

(70) Var. L. 5.97: Hircus, quod Sabini fircus; quod illic fedus, in Latio rure hedus, qui in urbe ut in multis A addito haedus “Hircus ‘goat’ which the Sabines call fircus. What is there a fedus, in rural Latium a hedus, is in the city, as is the case with many words, haedus with an added A.” On the attribution of these forms to
Sabine see Burman 2017:40–6. 

(71) Var. apud Vel. Long. (Keil 7.69.8). The fuller passage reads: ut testis est Varro a Sabinis fasena dicitur, et sicut s familiariter in r transit, ita f in uicinam adspirationem mutatur. similiter ergo et haedos dicimus cum aspiratione, quoniam faedi dicebantur apud antiquos; item hircos, quoniam eosdem aeque fircos uocabant. nam et e contrario quam antiqui habam dicebant nos fabam dicimus ‘As Varro testifies, fasena is said by the Sabines [for harena] an just as s commonly becomes r, so f changes into the related aspiration. In a similar fashion therefore we say haedos with h since they used to be called faedos among the ancients. Further hircos since they used to likewise call these same fircos. And in the opposite way we call fabam what the ancients called habam. Other f for h examples without dialect attribution: fordeum ‘barley’ (Quint. 1.4.14) = hordeum, folus ‘vegetable’ = holus, fostis ‘enemy’ = hostis, fostia ‘victim’ = hostia (last three examples all attributed to the antiqui at Paul. Fest. p. 74 L). 

(72) See Joseph and Wallace 1991a:84–93.

 
(73) Note that in Old Faliscan f (far ‘grain’ Bakkum 1, fileo ‘son’ Bakkum 471) and h (huti[c]ilom ‘vase’ Bakkum 1 < *ĝhuti-) are kept distinct. 

(74) The letter transcribed with F is in fact a backwards digamma. Other apparent examples of f for h in Praenestine are found in FELENA= Helena (CIL 1.2.566), FORATIA = Horātia (CIL 1.2.166) and and FELIOD(ORUS) = Heliodorus (CIL 1.2.1446). 

(75) The confusion of f and h seems to have been limited to initial position. First, there is some reason to think that h was not pronounced at all in medial position even in the Classical period. Second, there are no good examples of such an interchange. The forms trafere for trahere and uefere for uehere sometimes cited in this connection come from the work of a certain Apuleius Minor, De nota aspirationis p. 94 Osan. trahere pro trafere (attributed to Varro), p. 125 illi uefere Romani uehere protulerunt. But this work dates to the 10th/11th century and the forms are probably worthless grammatical creations on the basis of the initial-position f ~ h uncertainty. 

(76) Servius ad Aen. 7.695 and Ovid Fast. 4.73-4 mention that eponym of the Falisci was Hal(a)esus which Servius explains as a case of the Faliscan change of h to f. (Faliscos Halesus condidit. hi autem inmutato H in F, Falisci dicti sunt). Here the knowledge of a real linguistic phenomenon licensed a connection between the name of the Falisci and an unrelated name, cf. the Sicilian town (H)alaesa (Strab. 6.2.5) on the river (H)alaesus (IG 14.352). The place name Formiae is said by Servius in the same passage to have originally been Hormiae and Hormiae is given as the original form of the name by Strabo 5.3.6 (Ὁρμίαι λεγόμενον πρότερον δία τὸ εὔορμον), Festus, p. 73 L (oppidum appellatur ex Graeco, uelut Hormiae) and Pliny the Elder, Nat. 3.59 (Formiae, Hormiae prius dictae olim, sedes antiqua Lestrigonum). This idea, which may go back to Varro, is probably inspired by etymological fantasy. The clearest case of an artificial form is farreum said by Festus p. 73 L to be an ancient form of horreum ‘granary’ which was evidently created to support an etymological connection with far, farris ‘spelt’. On the phenomenon of “reconstructed forms” in the discourse of Latin grammarians see Zair 2019. 

References

Burman, Annie Cecilia. 2017. De Lingua Sabina. A Reappraisal of the Sabine Glosses. Ph.D. diss. Cambridge University. 

Hiersche, Rolf. 1965. “Der Wechsel zwischen anlautendem f und h im Lateinischen.” Glotta 43:103–18. 

Zair, Nicholas. 2019. “Reconstructed forms in the Roman writers on language.” Language & History 62: 227–46.