^{page 158 note 2} They are formed on the modified root *moi whence *moin, moen- and finally mun-.

^{page 158 note 3} A third meaning onus is recognized by the jurist Paulus (Dig. 50. 16. 18) and accepted by Lewis and Short who cite Cic. Verr. 2. 5. 51 and 52, Livy 25. 7. 4 and 27. 9. 9. But these are simply particular cases where the obligation was felt as a burden.

^{page 158 note 4} It is not questioned that munera ended as identical in meaning with munia, though it will be argued below that it was not so originally.

^{page 159 note 1} The general meaning of munifex is preserved in Pliny, N.H. 11. 234.Google Scholar The word is otherwise confined to military contexts. (In Dig. 50. 16. 18 munifices (Budé) should be read for munificos.) Vegetius, (Epit. Rei. Mil. 2. 7Google Scholar ‘Hi sunt milites principales, qui privilegiis muniuntur. Reliqui munifices appellantur quia munera facere coguntur) seems to take munera as ‘fatigues’ (cf. 2. 19 where they are listed) but Paulus/Festus p. 30 L ‘Beneficiari dicebantur milites, qui vacabant muneris beneficio; e contrario munifices vocabantur, qui non vacabant, sed munus rei publicae faciebant’, though recognizing this derivation, suggests that a soldier was munifex because he was fulfilling his obligations to the State; and this is almost certainly right.

^{page 159 note 2} This meaning is not found before Tacitus. ‘Munera aedilitatis’ in Cicero, Verr. 2.1. 14, refers to gladiatorial games, for which the use of munera has a different explanation. Vide infra.

^{page 159 note 3} Dumézil, , Mitra-Varuna, p. 74,Google Scholar attempts to fit this concept into a general picture of elements common to early Rome and early Indian society. But his methods have been criticized by Rose, in Journal of Roman Studies, xxxvii (1947), pp. 183 f.CrossRefGoogle Scholar

^{page 159 note 4} Cf. the Grammarians' distinctions. Keil, , Grammatici Latini, vii. 524. 16.Google Scholar (Cornelius Fronto) de diff.—munus quod amicus vel cliens vel libertus officii causa mittunt, vel munus gladiatorium, donum quod diis datur: vii. 119. 6 (Agroecius) donum dantis est, munus accipientis = Isidore diff. i. 360; so Cic. Clu. 34, Pis. 93, accipere munus. Cf. Meillet-Ernout ³, p. 749 ‘present que l'on fait (et non que l'on reçoit)’. But a ritual gift is both given and received.

^{page 160 note 1} Varro gives the two senses different derivations: L.L. 5. 179.Google Scholar The sense ‘gladiatorial games’ is probably derived not from the notable duties of the aedile and others (Meillet-Ernout ³, p. 749) but from the meaning ‘gifts to the dead’ (Schneider, K., Pauly-Wissowa, Suppl. Band iii. 761),Google Scholar or less likely from the gifts made to the people by the magistrate in thanks for his past and hopes for his future election.

^{page 160 note 2} MrWells, A. F., of University College, Oxford, once kindly explained the phrase munificat salute for me as a variant of donat salute and suggested that salute has the double meaning of ‘greeting’ and ‘salvation’.Google Scholar

^{page 160 note 3} Certum est suggests the rejection of a false alternative, which may have been given by Verrius Flaccus. These two are the only surviving instances of munis. But we cannot make any inference how rare the word was from the fact that no ‘extra-quotations’ of munis are made by Nonius Marcellus. Lindsay, (Nonius Marcellus' Dictionary of Republican Latin, p. 81)Google Scholar holds Uiat Nonius made such ‘extra-quotations’ only from those of his ordered sources that came after the one from which he drew the lemma. But he uses Plaut, . Merc. 105Google Scholar as an ‘extra-quotation’ for munia, a lemma derived from a source later than Uiat for die lemma munis, and should therefore have cited it also for munis. (See Lindsay, , Philologus, lxv, p. 442.Google Scholar) It seems therefore that citations were made haphazardly.

^{page 161 note 1} Professor Palmer suggests that Plautus may be using an old religious formula. Cf. Meillet-Ernout ³, p. 502, for the connexions of gratus.Google Scholar

^{page 161 note 2} But the derivation cannot be questioned. Compounds in -ceps fall into two well-defined classes, those compounded with caput which are declined -cipitis, and those derived from capio which are declined -cipis. In both cases the suffix is meaningful (see Walde-Hofmann, , i, p. 160Google Scholar and Meillet-Ernout ³, p. 172Google Scholar). The suggestion of Meillet-Ernout, based on Paulus/Festus, p. 65 L (deinceps), p. 115 L (manceps), that the suffix was sometimes passive is not well founded: manceps was originally active (Meillet-Ernout ³, p. 679Google Scholar), menceps is very doubtful (M. Hertz (Keil, , G.L. vol. 1Google Scholar) excludes it from Priscian 5. 66), and princeps and the analogous numerals of the sacra Argeorum (Varro, , L.L. 5. 45 ff.Google Scholar) are also active (Meillet-Ernout ³, p. 947).Google Scholar

^{page 162 note 1} Rubino, , Zeitschrift für Altertumswissen-schaft, 1844, 989,Google Scholar pointed out that hospites were free from munia, though Karlowa, , Romische Rechtsgeschichte, i, p. 288,Google Scholar plausibly suggested that this was only for short visits and that the hospes was privileged to take up a secure residence in Rome which involved the performance of munia. Mommsen dismissed the theory as impossible (Staatsrecht, iii, p. 231,Google Scholar n. 1 = Droit public, vi. i, p. 261, n. 1).Google Scholar

^{page 162 note 2} During the intervening period a man is presumably munis.

^{page 163 note 1} It is tempting to a non-philologist to explain moenia as a re-formation from munio a the sense ‘I perform my duties of labour’, for which cf. Lex coloniae Genetivae Juliae xcviiGoogle Scholar (Mommsen, , Staatsrecht, iii, p. 227). Thi would not affect the argument; but there are probably good arguments against this suggestion.Google Scholar

^{page 163 note 2} This is not the place to discuss these complicated but valuable sources. But it can be shown that they both derive from Verrius Flaccus, and they do not seem to be as corrupt or confused as has been supposed.

^{page 164 note 1} There seems no theoretical reason why a municeps should or should not be regarded as some sort of a Roman citizen, nor why his munia need be incompatible with full citizenship of another state. These are questions of fact, and their answers cannot be deduced from the meaning of the word.