MASTER NEGATIVE 
 
 NO. 93-81332-11 
 
MICROFILMED 1993 
 COLUMBIA UNIVERSITY LIBRARIES/NEW YORK 
 
 as part of the 
 "Foundations of Western Civilization Preservation Project" 
 
 Funded by the 
 NATIONAL ENDOWMENT FOR THE HUMANITIES 
 
 Reproductions may not be made without permission from 
 
 Columbia University Library 
 
COPYRIGHT STATEMENT 
 
 The copyright law of the United States - Title 17, United 
 States Code - concerns the making of photocopies or 
 other reproductions of copyrighted material. 
 
 Under certain conditions specified in the law, libraries and 
 archives are authorized to furnish a photocopy or other 
 reproduction. One of these specified conditions is that the 
 photocopy or other reproduction Is not to be "used for any 
 purpose other than private study, scholarship, or 
 research." If a user makes a request for, or later uses, a 
 photocopy or reproduction for purposes in excess of "fair 
 use, that user may be liable for copyright Infringement. 
 
 This Institution reserves the right to refuse to accept a 
 copy order if, In Its judgement, fulfillment of the order 
 would Involve violation of the copyright law. 
 
AUTHOR: 
 
 MARTIN, HENRY 
 
 TITLE: 
 
 NOTES ON THE SYNTAX 
 
 OF THE LATIN... 
 
 PLACE: 
 
 BALTI 
 
 DA TE : 
 
 1909 
 
 ORE 
 
COLUMBIA UNIVERSITY LIBRARIES 
 PRESERVATION DEPARTMENT 
 
 BIBLIOGRAPHIC MICROFORM TARGET 
 
 Master Negative # 
 
 Original Material as Filmed - Existing Bibliographic Record 
 
 Restrictions on Use: 
 
 f877*52 
 U364 
 
 •^mmmr'^'wr' 
 
 mi^- 'i -'i — awi - ' iu w I 1 >i|i»i' 
 
 ■•wwMpn 
 
 Martin, Henry, 1878- 
 
 Notes on the syntax of the Latin inscriptions found in 
 Spain ... Baltimore, J. H. Furst company, 1909. i 
 
 49 p., 1 1., 23^™. I 
 
 • 
 
 Thesis (ph. d.)— Johns Hopkins university. I 
 
 Vita. i 
 
 Library of Congress 
 
 O 
 
 9— 1906S 
 
 TECHNICAL MICROFORM DATA 
 
 REDUCTION RATIO: I/JL 
 
 FILM SIZE:__3_flm'_. 
 
 IMAGE PLACEMENT: lA (TIA) IB UB 
 
 DATE FILMED:___^liXl2i INITIALS_2^_^_<j 
 
 FILMED BY: RESEARCH PUBLICATIONS. INC WOQDBRIDGE. CT 
 
c 
 
 Association for information and Image Management 
 
 1100 Wayne Avenue, Suite 1100 
 Silver Spring. Maryland 20910 
 
 301/587-8202 
 
 Centimeter 
 
 III 
 
 2 3 
 li 
 
 4 5 6 
 
 iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiliiiiliiiiliiiiliiiiliii 
 
 WW 
 
 7 8 9 10 11 
 
 iiiliinlinilniiliiiiliiiiliiiiliniliiiil 
 
 12 13 
 
 iiiliiiiliiiilii 
 
 14 15 mm 
 
 ilimliiiiliiiil 
 
 I I 
 
 Inches 
 
 TTT 
 
 1 
 
 T 
 
 TIT 
 
 1.0 
 
 1^ 2.8 
 
 Mam 
 
 ■^ 111113 6 
 
 Li. IIIIM 
 
 U 
 
 •* u 
 
 1.4 
 
 2.5 
 22 
 
 I.I 
 
 2.0 
 1.8 
 
 1.6 
 
 1.25 
 
 TTT 
 
 MRNUFfiCTURED TO RUM STflNDRRDS 
 BY APPLIED IMAGE. INC. 
 
NOTES ON THE SYNTAX OF THE LATIN 
 INSCRIPTIONS FOiD IN SPAIN 
 
 fa" 
 11*. 
 
 l-i:- 
 
 w 
 
 
 BY 
 
 HENRY MARTIN 
 
 ai SDi0s?rrtation 
 
 PRESENTED TO THE BOARD OF UNIVERSITY STUDIES OF THE JOHNS HOPKINS UNIVERSITY 
 IN CONFORMITY WITH THE REQUIREMENTS FOR THE DEGREB OF 
 
 DOCTOR OF PHILOSOPHY 
 
 BALTIMORE 
 
 J. H. FURST COMPANY 
 
 1909 
 
*A \ 
 
 WTT^^ 
 
 ^^Vh^ 
 
 Columbia Winibtvaiip 
 
 LIBRARY 
 
 W i 
 
 ;i 
 
 COLUMBIA UNIVERSITY LIBRARIES 
 
 This book is due on the date indicated below, or at the 
 expiration of a definite period after the date of borrowing as 
 provided by the library rules or by special arrangement with 
 the Librarian m charge. 
 
 DATE BORROWED 
 
 mwi 5»*8 
 
 DATE DUE 
 
 DATE BORROWED 
 
 DATE DUE 
 
 C28(747) M100 
 
NOTES ON THE SYNTAX OF THE LATIN 
 INSCRIPTIONS FOID IN SPAIN 
 
 BY 
 
 HENRY MARTIN 
 
 3L SDfesfertation 
 
 PRBSENTED TO THE BOARD OF UNIVERSITY STUDIES OF THE JOHNS HOPKINS UNIVERSITr 
 IN CONFORMITY WITH THE REQUIREMENTS FOR THE DEGREE OF 
 
 DOCTOR OF PHILOSOPHY 
 
 BALTIMORE 
 
 J. H. FURST COMPANY 
 
 1909 
 
TABLE OF CONTENTS. 
 
 
 InTKODUCTION PAGE. 
 
 5-6 
 
 Concord 
 
 ,^. . 7-11 
 
 Adjectives, 7 ; Pronouns, 8 ; Verbs— Singular for Plural, 9 ; 
 
 Plural for Singular, 9. 
 
 The Cases 
 
 -_ . . 12-27 
 
 JSTommative Case— Nominative in Titles, 12 ; Nominative of the 
 
 Name, 12 : Genitive Case-Limiting Genitive, 13 ; Possessive 
 
 Genitive, 13 ; Partitive Genitive, 13 ; Genitive of Quality, 14 ; 
 
 Genitive with Adjectives, 15 ; Genitive with Verbs, 16:' 
 
 Dative Case-Dative with Verbs, 16 ; Mihi nomen est, 18 : 
 
 Accusative Case— Accusative with Verbs, 18; Ablative 
 
 Case-Ablative of Quality, 19 ; Modal Ablative, 19 ; Abla- 
 
 tive used Adverbially, 20 ; Ablative with Adjectives 20 ; 
 
 Ablative with Verbs, 21 ; Expressions of Time, 22 ; Local 
 
 Kelations, 24 ; Ablative Absolute, 26. 
 
 Interchange of Cases 
 
 • 27—30 
 
 Nominative and Genitive, 27 ; Nominative and Accusative, 28*; 
 
 Genitive and Vocative, 28 ; Genitive and Ablative,' 29 \ 
 
 Accusative and Ablative, 29. 
 
 Pronouns 
 
 30-31 
 
 Verbs. 
 
 _ 32-40 
 
 Tenses— Present for Future and Vice Versa, 32 ; Future Perfect, 
 
 32 ; Tenses of the Infinitive, 33 : Moods-Infinitive, 33 ;' 
 
 Indicative in Indirect Questions, 34 ; Subjunctive for the 
 
 Imperative, 34 ; Mood after Conjunctions, 35 : Impersonal 
 
 Verbs, 35 ; Kegimen of Verbs, 36 ; Compound Verbs, 38. 
 
 Prepositions 
 
 40-46 
 
 Adverbs 
 
 46-48 
 
 Conclusion 
 
 48-49 
 
INTRODUCTION 
 
 The subject of Latin inscriptions found in Spain is not an 
 untried field of research. In view of the rather complete work 
 upon this material recently published bj Carnoy,! a brief state- 
 ment in regard to the purpose of this pamphlet will be relevant. 
 It is not the object, at least the primary object, of this work 
 to solve any of the great problems that have so long engaged 
 the attention of students of language, such as those broached by 
 Grober, Sittl, Mohl and others. On the contrary, it is con- 
 fessedly syntactical, undertaken with a view to seeing what 
 contribution these inscriptions would make to Latin syntax as 
 it is at present worked out, what details a study of them could 
 add to the grammar of the language, and further what ex- 
 amples of rare usage could be added to those found in works 
 on syntax in case the testimony of Spanish inscriptions should 
 ever come to be embodied in such a treatise, and incidentally 
 to observe what light, if any, could be thrown on the relation 
 between the Latin of these documents and the Spanish language. 
 The difficulty of such a task owing to scarcity of material and 
 want of accurate dating, and the delicacy and care in inter- 
 pretation required by this matter have been comprehensively 
 stated by Carnoy in the introduction to his treatise ; they do not 
 need reiteration here. Wherever possible, comparison with the 
 results obtained by Pirson ^ for inscriptions of Gaul has been 
 made and to facilitate such a comparison a selection of topics 
 and a plan of arrangement essentially similar to his have been 
 adopted; for this no apology is needed. As far as possible 
 those points treated by Carnoy have been omitted, though this 
 material was for the most part arranged when his articles ap- 
 
 'Le latin d'Espagne d'apr^s les inscriptions. Bruxelles, 1906. 
 
 = La langue des inscriptions latines de la Gaule. Bruxelles, 1901. 
 
^ Notes on the Syntax of the 
 
 peared. In a few cases for the sake of symmetry points al- 
 ready noted by him have been retained. 
 
 The material dealt with is substantially that given by Carnoy, 
 with the exception that the inscriptions on amphorae, C. I. L., 
 Vol. XV, were not used, and with the addition of the Ephemeris 
 Epigraphica, [N'o. 9, which appeared subsequent to the publica- 
 tion of Carnoy's articles in the Museon. The works most fre- 
 quently cited and followed are those on Syntax by Kiihner, 
 Draeger, :N'eue and Schmalz and the work of Bonnet on the 
 Latin of Gregory of Tours. The following abbreviations may 
 require explanation: G-L.=Latin Grammar, ed. Gildersleeve- 
 Lodge, 1894; I. H. C.=Inscriptiones Hispaniae Christianae, 
 ed. Hiibner. The inscriptions cited from the 2d volume of 
 the Corpus are indicated by numbers simply. 
 
 Latin Inscriptions found in Spain, 
 
 COKCOED. 
 
 I. Adjectives. 
 
 An expression made up of a noun and some other word 
 joined to it by the preposition cum, in classical Latin is com- 
 monly treated as singular. A few cases are found in these 
 inscriptions where such a word group is regarded as singular 
 judged by the number of the verb, but as plural when modified 
 by an adjective. This construction is both natural and clas- 
 sical.^ I. H. C. 259, Sanxit cum coniuge . . . duobusque 
 natis . . . caventes. Parallel with this, however, appears a 
 construction occurring not so frequently in Cicero,, but found 
 often and with extended application in writers of the Augustan 
 Age, in which the modifier of a collective noun or its equivalent 
 is written in the plural, even though the verb may stand in the 
 singular.2 2071, Ordo . . . dedicatissimi d. d. ; 3730, 6004, 
 Sodalicium vernarum Isidem colentes ; I. H. C. 244, Qui in 
 hac aula dei ingreditur . . . deponant ingredientes. 
 
 Some of the examples of irregularity in agreement to be cited 
 in other connections show violation of the rules of gender in 
 addition to other peculiarities. At this point may be recorded 
 a few examples in which the departure from this principle seems 
 to be the primary feature. 3509, Valeria . . . se vivo fecit; 
 4248, Statuam . . . positam exomandum . . . censuer(unt) ; 
 1580, Conditus officio Graeca ser. hie. sita est; 4512, Propter 
 insignia bene gesti proconsulates omnes; I. H. C. 22^, Era 
 siscens quattus ; I. H. C. 379^, Cui . . . crustatam stringat 
 tortilis ansa latus. 
 
 In the majority of the cases just given noun and adjective 
 
 ^Kiihner, AusfUhrliche Grammatik der lateinischen Sprache. 1878, 2, p. 
 34, par. 14. 2. 
 *Kuhner, 1. c, 2, p. 16, par. 9.2. 
 
I 
 
 8 
 
 Notes on the Syntax of the 
 
 are separated by one or more words, yet neglect of concord may 
 occur even where they are in immediate contact as the follow- 
 ing will show: 2295, Adsiduae mater; 4310, Incomparabili 
 religionis eiiis; I. H. C. 335, Hac tumulo; I. H. C. 215, Pastor 
 suique obibus . . . repellit mimdi delicia. 
 
 II. Pkonouns. 
 
 J 
 
 ^^pi^e^ni^yfcje^pect^ glaring inconsistencies in the agreement 
 of the relative with its antecedent appear. Bonnet ^ has re- 
 corded similar constructions found in the Latin of Gregory of 
 Tours and is probably correct in explaining one phase of it,- 
 at least, as due to the similarity in sound between the different 
 case forms of the relative in rapid pronunciation. In the ex- 
 amples cited below it will be seen that little difference is made 
 V between quern and quam, and even between qui and quae,^ so 
 that even in these few examples there seems to be definitely 
 foreshadowed that leveling of the forms of the relative so charac-^ 
 teristic of the Eomance languages.^ For a similar list made 
 up from inscriptions of Gaul see Pirson.^ — 2211, Tesseram 
 offerimus . . . quem . . . iubemus suscipi; 5393, Memoria 
 . . . quem fecit; I. H. C. 258, En quem cemis . . . Tarasia 
 Christo dicata; I. H. C. 403, Saturina qui vixit; I. H. C. 519, 
 Eulalia . . . qui passa est. 
 
 In the case of personal pronouns practically no irregularity 
 occurs. Worthy of mention, however, are two examples in 
 which the pronoun fails to be adapted in number to the person 
 referred to, but occurring as it does in a formula the error is a 
 mere inadvertence. 1993, Politice an. L, Crysidae an. V, 
 Pusinnicae an V, h. s. s., s. t. t. 1. ; also 5536, see p. 9. 
 
 *Le latin de Gr^goire de Tours, p. 501. 
 'I.e., p. 499. 
 'Carnoy, I.e., p. 269. 
 
 * Meyer-Liibke, Graminaire des langues romanes, 2, pp. 143-144. 
 ' 1. c, pp. 158-159. 
 
 Latin Inscriptions found in Spain. 9 
 
 III. Verbs. 
 
 1. Singular for Plural. 
 
 (a) With compound subject. — In classical Latin two or more 
 subjects in the singular connected by et or other co-ordinating 
 conjunction might be followed by a verb in either singular or 
 plural number according as they were regarded as acting inde- 
 pendently or in concert.^ In these inscriptions the construc- 
 tion is of frequent occurrence and with such a subject the sin- 
 gular is so uncommon that the few examples of it are worth 
 recording. 5393, Quem fecit pater et mater et Leucadius; 
 I. H. C. 230^ (p. 120), Eudesindus episcopus et Petrus abbas 
 me iussit fieri. In a certain number of inscriptions real irreg- 
 ularity is caused by the employment of formulas in an ab- 
 breviated form, so that the proper connection between them 
 and the rest of the sentence was lost. In these cases, no doubt 
 the graver applied the traditional and well-worn formulas with- 
 out thinking to adapt them to the number of persons mentioned. 
 3344, Q. Lucretius Silvinus . . . et Q. Lucretius Silvanus pi 
 in sues h. s. e. ; 3658, L. Fabius Proculus . . . Vitoria Omul- 
 lina . . . h. s. e. s. ; 5536, M. Sempronio M. 1. principi, Sem- 
 p(r)oniae L. 1. Secundae, Luriae C. f. Faustillae . . . h. s. e. 
 
 S. v. Xi. 1. 
 
 (h) With plural subject. — It is possible to explain most of 
 the examples under this caption by saying that the subject, 
 though plural, is regarded as a mass or entity and so is fol- 
 lowed by the singular, or else separated from the verb by a 
 phrase and in this way the thought is not accurately maintained. 
 5439 i-^iij Quicumque pontif (ices) quique augures . . . lectus 
 co-optatusve erit ; 6278 ^^, Ut ea pretia . . . pertineat ; I. H. C 
 396, Datum est . . . in auro soledos III. 
 
 2. Plural for Singular. 
 
 In Spanish inscriptions the examples of two nouns connected 
 by -ve both occur in one inscription and both are followed by 
 
 *G-L., Latin Grammar, par. 285, Rem. 1; Kuhner, 1. c, 2, par. 14. 1; 
 Pirson, I.e., p. 161. 
 
10 « 
 
 Notes on the Syntax of the 
 
 plural verb.^ 5439 -•^•^, Eamque pecimiam II vir prae- 
 f(ectnsve) dandam . . . curanto; 5439 -•-•^^, Itqiie ... II vir 
 aedil(is)ve dinioliendiim curanto. Examples of irregularity, 
 however, are seen and each is capable of individual explana- 
 tion. 6278 ^, Procurator qui provinciis praesidebunt ; prob- 
 ably brought down by mistake from his qui . . . praesidebunt 
 V. 41 ; 6278 ^^, Lanistas . . . negem sibi copiam . . . prae- 
 bendas esse; where the plural may be taken from lanistas; 
 I. H. C. 157, Arcadius vixit annos VII et requieverunt in pace 
 . . . et fuerunt servi . . . Vinccnti, etc., where the plural idea 
 seems to be suddenly adopted. One of the most interesting 
 examples here is the following: 3570, M. Sempronius Hymnus 
 suo et M. Semproni Reburri fili sui nomine . . . restituerunt. 
 The sentence is logically correct, since the author had two per- 
 sons in mind. In reality a noun combined with a prepositional 
 phrase, a modifier of the verb, is regarded as plural. Similar 
 to this is a case in which the structure of the sentence was 
 changed after the verb was reached, or else the subject was 
 completely lost. I. H. C. 390, Hie requiescit . . . cui decies 
 . . . (per) metas regendi monacos incubuere. 
 It has already been mentioned (p. 7) that a subject in the 
 singular modified by a cum phrase might take a plural verb, 
 but as a matter of fact is seldom treated as plural in the best 
 writers.- For the entire body of Gallic inscriptions, even, 
 Pirson ^ has recorded but a single example of the plural with 
 such a subject. In Spanish inscriptions, too, this type of 
 subject is for the most part treated as singular. The following 
 cases of plural may be noted: 3298, Fausta cum filio . . . hie 
 sit. sunt; I. H. C. 334, Ut domus haec cum habitatoribus . . . 
 florescant. To these may be added one other example, in which 
 this combination is qualified by a word in the plural in apposi- 
 tion with the phrase and the plural verb is in consequence 
 
 Latin Inscriptions found in Spain. 
 
 11 
 
 natural and normal.^ I. H. C. 465, Offerunt fjimuli dei Ade- 
 fonsus . . . cum coniuge. Especially was the singular in 
 vogue when such a compound subject represented a difference 
 of persons.^ The one example in Spanish inscriptions is at 
 variance with this. I. H. C. 283, Ego Sanccius . . . cum 
 uxore mea . . . iussimus. 
 
 ^Klihner, 1. c, 2, p. 34. 2 (end). 
 = Kuhner, 1. c, 2, p. 34, par. 14. 2. 
 
 4 
 
 ;1 
 
 ^Kiihner, I.e., 2, par. 14. 1. 
 
 *G-L., 1. c, par. 285. 3, Rem. 2; Draeger, Historische Syntax der latein- 
 ischen Spraehe, 1, p. 178, par. 105. 
 •I.C., p. 163. 
 
12 
 
 Notes on the Syntax of the 
 
 THE CASES. 
 
 !N^OMiNATivE Case. 
 
 Nominative in Titles: — The l^ominative is used in inscrip- 
 tions of various kinds that imply action or character.^ This is 
 in reality the Nominative of the title. The only apparent ex- 
 ceptions to this principle occur when monuntentiim, dis mani- 
 hus, or some other word or phrase thoroughly familiar is for 
 the sake of brevity omitted. In this event the title is to all 
 appearances in the Genitive or Dative case. This is common 
 in inscriptions everywhere and no examples need be cited. The 
 titles in a single inscription, however, in which I^ominative and 
 Genitive seem to be used side by side deserve notice. They are 
 given without comment, since Hiibner's note on the inscription 
 seems complete. The titles are as follows: 5181, I. Centesimae 
 argentariae stipulationes. II. Scripturae praeconii. III. Balinei 
 fruendi. IV. Sutrini. V. Tonstrini. VI. Tabemarum ful- 
 loniarum. VII. Scripturae scaurariorum et testariorum. VIII. 
 Ludi magistri. IX. Usurpationes puteorum sive pittaciarium. 
 
 Nominative of the Name: — It was in the poets of late Latin 
 that instead of an oblique case we find the Nominative as the 
 case which simply names a person or thing, as in the example 
 " Lazarus hie habuit nomen " cited by Schmalz.^ Probably 
 the number of instances of this type for the entire literature 
 is far from large. In view of this, therefore, the one example 
 found in these inscriptions is amply worth mentioning and 
 recording. I. H. C. 12, Venantia nomen . . . gesi. 
 
 * Gildersleeve, Syntax of Classical Greek, par. 5. 
 
 ^Lateinische Syntax, Miiller's Handbuch der elassischen Altertumswis- 
 senschaft, II, 1900, p. 227, par. 43; see also Draeger, 1. c, 1, 390, par, 172. 
 
 
 Latin Inscriptions found in Spain, 13 
 
 Genitive Case. 
 
 Limiting Genitive: — The sphere of the Genitive in its 
 simplest form is considerably widened in the language of in- 
 scriptions, so much so that it often appears to indicate ignorance 
 on the part of the writer of the idiomatic Latin turn or to be 
 his method of expressing an idea in the fewest possible words 
 without reference to clearness. The result may be a heaping of 
 Genitives and a good example is the following: 5260, Valeria 
 Avita aram tauriboli sui natalici redditi d. d., where Hiibner 
 explains thus: ara posita propter taurobolium deae redditum 
 die natali Avitae. Under this head, too, may be placed a 
 strange use of the Genitive with the words tempus and vetustas 
 in which the latter in the Genitive is employed as a descriptive 
 term of the former, which then virtually becomes redundant. 
 4756, Vias et pontes tempore vetustatis conlapsos; so also 4826, 
 4853. But 4788, Temporis vetustate conlabsos. 
 
 Possessive Genitive: — The Genitive is found in classical 
 Latin after the preposition ad, rarely after others, the word 
 templum being omitted.^ The only examples approaching this 
 in Spanish inscriptions are extremely uncertain, but if correct 
 mark an advance over the above principle. 2146, Si quis 
 titulum ad mei nominis astiterit. Here Hiibner thinks me- 
 moriam is the omission, Mommsen regards the preposition as 
 simply post-positive. I. H. C. 472, Episcopus de See Marie 
 feci. 
 
 Partitive Genitive: — There may be noted here the Genitive 
 dependent upon neuter adjectives in the singular or plural em- 
 ployed as nouns, a construction sparingly used by classical 
 writers, but extended by Livy and later authors.^ Not many 
 examples occur and these few are found in late Christian in- 
 scriptions. I. H. C. 385, Ut numquam sentiat seba Gehennae ; 
 LH. C. 218^ Per devia mundi; L H. C. 218% Qui temnens 
 fluida conscendit lucida caeli ; I. H. C. 385, Intima qui penetras 
 
 *Kuhner, 1. c, 2, p. 175. 6. 
 
 'Draeger, I.e., 1, p. 453. 4; Schmalz, I.e., p. 237. 62. 
 
14 
 
 Notes on the Syntax of the 
 
 Latin Inscriptions found in Spain. 
 
 15 
 
 cunctorum. Such expressions as saevissimus saevorum are not 
 real Latin, but are found in those authors who are most in- 
 fluenced by the Greek. The type selected is not altogether 
 fortunate, since its parallelism with the following is not exact. 
 However, somewhat like this are a number of phrases in these 
 inscriptions with the difference that they are made up of nouns 
 rather than of adjectives. They appear only in Christian in- 
 scriptions, where the influence of the Vulgate is manifest and 
 are without doubt Hebraisms in origin. There may be, how- 
 ever, an element of the popular in them.^ I. H. C. 49, 483, 
 Saecula saeculorum ; I. H. C. 218, Presta deus deorum. 
 
 Genitive of Quality: — While used at all periods of the lan- 
 guage to express quality, the Genitive was in especial favor 
 in popular speech, where it seems to have been used somewhat 
 to the exclusion of the Ablative.^ Bonnet ^ and Pirson ^ have 
 reached this conclusion in regard to the Latin of Gregory of 
 Tours and of Gallic inscriptions respectively. The fact is 
 noticeable in Spanish inscriptions, also, where it is used to such 
 an extent that the Ablative may be said to be of comparatively 
 rare occurrence. The following list, while by no means com- 
 plete, is intended to convey some idea of the extent and sphere 
 of the Genitive of Quality, with which may be compared a 
 briefer list given under the Ablative of Quality, p. 19: 131, 
 Deo Endovellico praesentissimi ac praestantissimi numinis; 
 4061, Theopompo . . . aedilici iuris; 4146, Uxori rarissimi 
 exempli; 2436, Feminae incomparabilis pietatis; 4315, Fac- 
 tionis Venetae Fusco sacravimus aram; 1699, Alumna priscae 
 pietatis; I. H. C. 362, Obiit felicis memoriae Leander. We 
 may further consider as Genitive of Quality a number of con- 
 structions with the Genitive which are used with some frequency 
 and which may be grouped as follows : 
 
 1. Tribal origin, domicile, etc. 1678, luliae G. f. . . . 
 domus Augustae; 365, Valerius Avit. . . . de vico Baedoro 
 
 »Schmalz, 1. c, 237. 62, Rem. 3; cf. Petron. 37, Habet nummonim 
 niimmos; 43, Olim olionim ( Friedliinder's note). 
 » Schmalz, 1. c, p. 236, par. 61. » 1. c, p. 548. * I. c, p. 175. 
 
 gentis Pin ton; 829, L. Domitius T. f. Gal. Vetto Otobesani; 
 I. H. C. 249, Concessum est sancto Salvatori Ovetensis sedis. 
 This is especially frequent in one large class of names mostly 
 foreign in origin and with Genitive plural in -um such as: 2708, 
 Cassio Corovescum ; 3082, Aplondus Dagencium M. f.-^ 
 
 2. Campaigns one has served, number of years one has lived, 
 or time in general. 1126, D. M. S. L. Valeri nepotis militis 
 stipendiorum VII annorum XXX; 4383, M. luvento incre- 
 mento . . . mensium IX; 3564, G. Lol(i)us Kufus annorum 
 XXXIV. Probably the only certain example of the Ablative 
 in expressions of the kind is : 3471, Herennia mortua est aetate 
 XV an. As in Gallic inscriptions, so in those of Spain, the 
 Genitive could be employed with the verbs ohire, recedere and 
 the like, and finally even with the verb vivere. In the case of 
 all of these the Genitive is of rather unusual occurrence.^ 3566, 
 Qui obit (us est) annorum XIV; 3297, Marito . . . qui vixit 
 annorum XXXX; I. H. C. 136, Recesset Tale . . . annorum 
 XLII; L H. C. 171, Occurrit Aniator etatis sue XLIII. 
 
 It is plain that the use of the Genitive discussed in both the 
 above captions marks an extension beyond classical bounds and, 
 as Pirson ^ suggests, it may have been influenced by the classical 
 Genitive of Quality. One would be safe in assuming further 
 that the beginnings of the Genitive discussed in the latter were 
 already laid in Latin of the Augustan Age, for in Varro ^ then 
 in Livy and others is found maiores triginta annorum, to which 
 Schmalz ^ traces the expressions natus decern annorum, cum 
 esset annorum decern, vixit annorum quattuor of Sulpicius 
 Severus. 
 
 Genitive with Adjectives: — In addition to its use with cer- 
 tain adjectives that regularly require the Genitive in classical 
 Latin is found its use with several others in such a way as not 
 to meet the requirements of standard prose. 
 
 Doctus, — 4313, Tib. CI. ApoUinaris . . . artis medicine doc- 
 tiss(imus). 
 
 ^ ITiibner, Monuraenta Linguae Ibericae, p. cxxxvii. 
 
 = Pirson, 1. c, p. 178. ^ 1. c, p. 177. 
 
 * De Re Rustica, 2. 7. 1. ' 1. c, p. 230. 61, Rem. 4. 
 
16 
 
 Notes on the Syntax of the 
 
 Fidelis. — Here is usually found the Dative or an equivalent 
 construction. In the example below, fidelis evidently has the 
 force of a noun and the Genitive is therefore used. I. H. C. 
 117, Maria fidelis Chr(ist)i in vita sua. 
 
 Studius, — A word not appearing in the lexicons. It probably 
 follows the construction of adjectives of desire and in this case 
 studiosus would naturally be thought of and the Genitive would 
 not be abnormal.^ 6109 (Nee) ip(s)i stud (i) us artis. 
 
 Unctus, — The normal construction after this participle is the 
 Ablative, cf. I. H. C. 463, Unctus crismate. But a strange use 
 of the Genitive with this word is seen in a single inscription of 
 very late date. One would think of a change of declension, but 
 there is nothing to render this certain.^ I. H. C. 464, Crismatis 
 uncta. 
 
 Genitive with Verbs: — The Genitive with verbs may be dis- 
 missed with brief mention. The classical construction with 
 liherare is Accusative of the person and Ablative of the thing. 
 Livy, however, seems to be the first to employ with it a Geni- 
 tive of the thing as in the phrase voti liherare; so also culpae 
 liherare,^ As cited in the example below, the Genitive is prob- 
 ably dependent upon the preceding noun rather than upon the 
 verb following, but even in that case it seems worthy of record- 
 ing as unusual. 1957, L. Porcio . . . qui . . . d(edit) ut 
 aes alien (um) rei p(ublicae) . . . lib(eraret). 
 
 Dative Case. 
 
 Dative with Verbs : — The most interesting cases of the Dative 
 observed in these inscriptions are connected with its construc- 
 tion with verbs and hence they may be mentioned first. 
 
 Dimittere. — Noteworthy is the Dative with dimittere, as seen 
 in I. H. C. 285, 286, 523, Dimittat ei deus. Here the com- 
 plete expression may have been dimittat ei peccata somewhat as 
 in I. H. C. 31, Quinigiae in hoc loco quiescentis omnia peccata 
 
 ^KUhner, I.e., 2, p. 319 (a). 'Bonnet, I.e., p. 364, n. 8. 
 
 •Livy 5. 28. 2; Draeger, 1. c., 1, p. 487; KUhner, 1. c., 2, p. 338. 7. 
 
 Latin Inscriptions found in Spain. 
 
 17 
 
 dimitte. In the examples just given dimittere appears to have 
 both the meaning and construction of a verb of pardoning. 
 
 Precari. — A verb which in the literature is transitive may 
 by a slight difference in the shade of meaning become intransi- 
 tive and as such take the Dative. Such a verb is precari, which 
 in one inscription seems to have taken on the meaning of " call 
 out to," '' pray to " and in this sense may legitimately take the 
 Dative. 5186, Si sensus er(it) meae quietis qu(a)e lasso 
 tibi dulcius precabor. 
 
 Sociare. — The Dative with sociare occurs first in Vergil.^ I. 
 II. C. 158, Ce(ti)bus s(an)c(t)orum merito sociatus resurgam. 
 
 Dative with Verbs of fighting: — ^AVith such verbs as certare, 
 pugnare, etc., the Dative is poetical and on this ground is justi- 
 fied in one inscription.^ 4315, Laudem cursus meruisti; cer- 
 
 tasti multis. 
 
 Dative with Verbs of joining and the lihe: — In the case of 
 a certain class of verbs, chiefly of joining, and the like, either 
 simple or compounded with cum, the best classical usage was 
 cum and the Ablative. More rarely the Dative was used. With 
 the past participle of these verbs the condition is described by 
 Draeger^ as one of wavering. In Spanish inscriptions they 
 appear almost exclusively with the Dative, showing that they 
 had moved away from the classical idiom and had allied them- 
 selves with later usage.^ 1399, Omnis ut aetas sanguine me 
 iunctam crcderet esse sibi ; I. H. C. 34^, Quod matris tumulo 
 iungeris; I. H. C. 218^, Amore quo tibi coniunctus mansit 
 . . . earns; I. H. C. 255, His pariter co(n)iunctae sunt; 
 I. H. C. 213, Atletis iungitur rite celestibus; I. H. C. 455, 
 Ethcreis iungatur sorte beata locatus. To these may be added 
 two others with the Dative, connumerare, which is not classical, 
 and coaequare, which is seldom used in that period. I. H. C. 
 409, Connumerandus Sanctis ministrabit antistes; I. H. C. 
 
 ^ Draeger, 1. e., 1, p. 426. 
 
 *Hor., Odes 1. 3. 13; Draeger. 1. c., 1, p. 406, par. 184. 
 
 "I.e., 1, p. 425. 
 
 * Bonnet, 1. c., pp. 541 flf. ' 
 
 |ii' 
 
18 
 
 Notes on the Syntax of the 
 
 Latin Inscriptions found in Spain, 
 
 19 
 
 390, Qui cursum . . . peregit . . . Pauli Antonique meritis 
 quoaequandiis in aethera. 
 
 Dative with Verbs of taking away: — The Dative not only 
 occurs with such verbs as eripere, but its use is further extended 
 to several verbs not ordinarily so construed in classical Latin. 
 It is perfectly regular in the following: 3475, Parentibus eri- 
 puit fortuna ... me; Eph. Ep. 8. 194, Casus me eripuit 
 matri e manibus ; I. H. C. 247, Quisquis auferre presumpserit 
 michi. But not so with suhducere, privare and possibly de- 
 ducere as in those below: 1425, Huic dono . . . heres XX 
 non deduxit; 1293, Subductum primae Pyladem haec ara 
 iuvcntae indicat; I. H. C. 165, Quern sibi non sobu . . . priva- 
 bit trans (itus is)te. 
 
 Mihi nomen est: — The well-known Latin idiom mihi nomen 
 est has to a large extent disappeared to be replaced by less 
 complex forms. For the most part periphrases are resorted to, 
 especially in the metrical epitaphs, or else a direct turn like 
 English " my name is, etc.," is employed. 1235, N'ome fuit 
 nomen; 3501, Huius nomen Salviol(ae) fuerat; 1235, Haesit 
 nascenti Cusuccia ; I. H. C. 12, ISTomen Venantia in saeculo 
 gesi. 
 
 Accusative Case. 
 
 The constructions with the Accusative case were less com- 
 plicated than those with the other oblique cases, were easier for 
 the masses to use and one would expect them, therefore,, to be 
 less subject to radical departure from best usage. Such is the 
 case in these inscriptions, where consen^atism in the use of the 
 Accusative is displayed and little out of the ordinary can be 
 recorded in its treatment. The employment of a direct object 
 with certain intransitive verbs, a phenomenon familiar in late 
 Latin, occurs to some extent, thereby confirming the popular 
 origin of these inscriptions. To those given below add others 
 cited by Camoy.* 
 
 Blandiri,—!. H. C. 214, Blandensque corda pleviu(m). 
 Carere, — 4427, Dulcem carui lucem. 
 
 ^l.c, p. 271; Bonnet, I.e., p. 533; Pirson, I.e., p. 170. 
 
 Contingere, — This growing disposition to convert intransitive 
 into transitive verbs is further exemplified by the verb co7i- 
 iingere, Not at all in classical Latin and rarely at a later 
 period does this verb take an Accusative as used below. 4514, 
 Quos honor . . . contigerit. 
 
 Manere. — Although it has the intransitive meaning of " re- 
 main," yet it takes the Accusative in one inscription and ap- 
 pears to be construed like manere in the phrases mors manet, 
 exitus manet, 1399, Cum me mea vita maneret. 
 
 Ablative Case. 
 
 Ablative of Quality : — It has been observed under the head of 
 the Genitive that this case tends to be used to the exclusion of 
 the Ablative where the quality of a thing is to be expressed. 
 It remains only to cite a few other examples, of which there 
 are not many, and to say that a large proportion of these, if 
 all were collected, would be found to appear in the Christian 
 epitaphs, which are often mechanical and imitatory in charac- 
 ter. 172, Qui . . . hostili animo fuerint; 6087, M. Aurel. 
 M. f . Emona patria ; I. H. C. 149, Eafila ... fide probata ; 
 I. H. C. 455, Ore modestus . . . mente quieta. 
 
 Modal Ablative :— The one example to be considered here is 
 an interesting case of extension due to analogy. The familiar 
 classical construction with the adjective invitus was the Ablative 
 Absolute, se or eo invito. It is never found with the Genitive 
 of a pronoun, as invito eius below. Forcellini ^ cites one ex- 
 ample of it with the Genitive of a noun,^ but this is by no means 
 a parallel. The case we are dealing with is simply an extended 
 use of invitus by analogy to the phrases iu^su eius, monitu eius, 
 hortatu eius and others.^ 5439 ^•^•^^ Dum ne invito eius opera 
 exigatur qui . . . erit. 
 
 ^Totius Latinitatis Lexicon. 
 
 =^Ulpian, Dig. 16. 3. 11, Credidit dominum non invitum fore huius solu- 
 tionis. 
 
 ' Draeger, 1. c, 1, p. 538. 
 
20 
 
 Notes on the Syntax of the 
 
 Latin Inscriptions found in Spain, 
 
 21 
 
 Ablative used adverbially :^ After the verbs habere, putare 
 and others, active and passive, instead of a predicate Accusa- 
 tive, a construction with the prepositions in and pro with such 
 words as locus, numerus, etc., followed by a Genitive could be 
 used in the literature.^ Further the Ablative of numerus was 
 irequently employed as a sort of Ablative of Specification 
 after numerals just as pondo with weights. Something like 
 what appears to be a cross between these two usages occurs here, 
 where numero appears not with a numeral, but with the adjec- 
 tive alius and would seem to be a free use of the word in this 
 connection developed out of its use with numerals. It is, there- 
 fore, virtually equivalent to an adverb.- I. H. C. 140, Ilic 
 sunt reliquiae numero sanctorum; I. H. C. 140, Et aliorum 
 numero sanctorum. 
 
 Ablative tvith Adjectives:— The Ablative with the following 
 adjectives exhibits peculiarities worthy of note : 
 
 Compos.—In good Latin the Ablative with compos is rarelv 
 found.3 5965, Voto sum compos. 
 
 Expers~lts use with the Ablative is confined chiefly to old 
 Latm.-* 172, Me . . . expertem patria incolumitate fortunis- 
 que omnibus faxint. 
 
 . P^enw5.— As in classical Latin, pjcnus is seen both with Geni- 
 tive and Ablative with no seeming difference. 761, Templum 
 . . . superis et Caesare plenum; 3633, Morumque" bonorum 
 pleno; L H. C. 362, Fide plenos; L IL C. 219, Plenisque 
 dierum. This result was to be expected with an adjective like 
 plenus; the real reason, then, for recording this at all is its pos- 
 sible construction in one example where it appears to take both 
 Genitive and Ablative and that in the same phrase. :Momm- 
 sen (cf. his note on the inscription) thinks some such word as 
 insigms upon which the Ablative depended has not been pre- 
 served m the reading. Hiibner, however, claims that the stone 
 
 ; ^raeger, I. c, 1, p. 385 (d). = Cf. Indices I. H. C, p. 155, col. 1. 
 
 Kuhner, 1. c, 2, p. 284. 7; Draeger, 1. c, 1, p. 476 (b) 
 *Kuhner, I.e., 2, p. 325, Rem. 6; Draeger, I.e., 1, p. 560 (f) ; Holtze 
 Syntaxis priseorum Scriptorum Latinorum, 1, p. 121. 
 
 has been preserved intact and that the reading as we have it is 
 correct. 3738, Pietate, iustitia, fortitudine et pleno omnium 
 virtutum. 
 
 Furus. — Cited with the Ablative without the preposition only 
 from Horace.^ An example can be added from these inscrip- 
 tions. I. H. C. 456, Nobilis mundoque purus. 
 
 Ablative with Verbs: — Certain verbs also in Spanish in- 
 scriptions are followed by constructions with the Ablative not 
 altogether normal and may be conveniently grouped here. 
 
 Conglutinare. — After this verb a construction with cum and 
 the Ablative is the normal one. We may cite one example of 
 the simple Ablative from Spanish inscriptions. I. H. C. 239, 
 Conglutinans se norma monastica. 
 
 Esse. — A strange use of the Ablative is its apparent con- 
 struction with esse, as if it were a verb of action, and hence 
 the usage must be construed as a kind of Ablative of Means or 
 Cause. Eph. Ep. 9. 176, Hoc fuit j)rovidentia actoris ut . . . 
 mitteres illam. 
 
 Perire. — After perire in classical Latin might be used a 
 noun in the Ablative denoting the cause or circumstances of 
 the action. For example, the Ciceronian phrase fame perire. 
 In Christian inscriptions this verb is followed by a noun morte, 
 of cognate meaning, which at first sight appears redundant and 
 analogous to the phrase cited above, but on closer examination 
 reveals the influence of the Vulgate and seems to convey the 
 paradoxical idea " he did not perish in death." L H. C. 218, 
 ^ec morte periit sed vibit sede perhenni. 
 
 Uti, — In a Christian epitaph of late period, utor is followed 
 by the relative adverb unde, which is, therefore, equivalent 
 to a construction with the preposition de. One or more similar 
 examples may be cited from early Latin.^ I. H. C. 255, De 
 pane . . . unde in cena usus est. 
 
 Yendere. — Here too may be mentioned a remarkable con- 
 
 * Sat. 2. 3. 213; Draeger, 1. c, 1, p. 500. 
 
 ^Ter., Adel. 981, cf. Parry's note (5. 9. 24.). Hine dederis 
 utatur. 
 
 unde 
 
22 
 
 Notes on the Syntax of the 
 
 struction with the verb vendere. In standard prose this and 
 kindred verbs took only pluris and minoris, Genitive of in- 
 definite price. Indeed with regard to plus no Dative nor Abla- 
 tive singular is ordinarily given by grammarians. The ex- 
 amples cited in works of syntax show few or no cases of the 
 Ablative of plus with a verb of buying or selling.^ Several 
 well-defined cases of plure marking price occur, all in one in- 
 scription of between the years 176-180 A. D. It is probably 
 an extension by analogy to the words magno, plurimo, parvo 
 and the like regarded as neuter substantives, so that finally in 
 popular Latin the expressions caro, vili emere could be used.^ 
 6278 % Neque singulatim aliquem rei gladitoriae causa vendat 
 plure, etc. ; occurs again 1. 58. 
 
 Expressions of Time. 
 
 By reason of great latitude of use, expressions of time merit 
 distinctive treatment. In Spanish inscriptions expressions de- 
 noting time alternate in the Genitive, Accusative and Ablative 
 cases. As has already been shown, the Genitive in this rela- 
 tion appears to be in origin, at least, the Genitive of Quality, 
 but finally comes to express duration of time with verbs side 
 by side with the Accusative and Ablative. The fact that the 
 Ablative is often used parallel with the Accusative and even 
 tends to supplant it in giving the age of persons, is a phe- 
 nomenon common to all inscriptions and need not be dwelt upon 
 here. It ought to be said, however, that this statement applies 
 principally to pagan inscriptions. In Christian inscriptions 
 the Accusative tends to reassert itself to such an extent that the 
 following almost complete list with the Ablative may be made 
 out for them: I. H. C. 71, 72, 73, Vixit annis ; I. H. C. 158, 
 Annis sexa(ginta p)eregi; I. IT. C. 330, 342, 367, Vixit 
 annis XIX. To these must be added a few that are placed 
 
 ^G.-L., I.e., par. 90; Kiihner, I.e., 2, p. 335, par. 5; Draeger, I.e., 1, 
 p. 464, par. 201, and p. 567, par. 247. But cf. Neue, Formenlehre der 
 lateinischen Sprache, 2, par. 71. 
 
 » Schmalz, 1. c., p. 249, par. 87 ; but cf . Carnoy, 1. e., p. 47. 
 
 Latin Inscriptions found in Spain. 
 
 23 
 
 below. It is not at all rare to find the Accusative and the Abla- 
 tive standing side by side in the same expression of time, thus 
 confirming their practical identity to express duration of time.^ 
 5443, Vixit annis XXXV, mens. V, dies XXVIII; 5144, 
 Vixit annis LII, mens I, dies V ; I. H. C. 139, In annibus 
 viginti duos; I. H. C. 215, Quattuor denis et duo annos; 
 I. H. C. 275, Duo et tribus mesibus. 
 
 In giving dates the classical construction a. d. hal. Ian. or 
 pridie nonas Ian. appears as follows : 
 
 1. Ante diem is entirely omitted while the adjective form 
 lalendas, etc., is still retained in the Accusative case, a method 
 common enough in classical Latin and is here given simply for 
 completeness. I. H. C. 3, Kequievit in pace ... XI kal. ^o- 
 
 vembres. 
 
 2. The forms die, sub die, take the place of a. d., etc., and 
 what follows is still put in the Accusative case. This is very 
 frequently used. In rare instances the Genitive is found.- 
 I. II. C. 46, Kecepta in pace die IV idus lanuarias ; I. H. C. 
 362, Die II idus Martias die II nonas Apriles; I. H. C. 396, 
 Sub die sexto id (us) ]Srovembri(s). Prepositions are but 
 sparingly used. I. H. C. 147, lustus ... in aera, etc.; so 
 I. H. C. 252, 253, 254; I. H. C. 220, Septem de kalendis . . . 
 Aprilis; I. II. C. 267, In ns Fbrii; I. H. C. 95, In novissimo 
 die; I. H. C. 472, In S(a)n(c)ti Micaeli(s) die. 
 
 Here may be called attention to peculiarities noted in a few 
 individual inscriptions. In the expression I. II. C. 143, Anno 
 sexcentum decies era nonagesima novem, sexcentum decies 
 seems to be a bold substitution for sexies centum. Die and pridie 
 are found side by side in one inscription and the whole expres- 
 sion seems to be the simple equivalent oi a. d. I. H. C. 316,. 
 Kequievit in pace ... die pridie kal. Februarias. The omis- 
 sion of any word to indicate time leaving an attribute with 
 which it is to be supplied is noteworthy in one example. 59, 
 Vivas pluribus (i. e. annis). Here, too, may be noted the 
 
 iPirson, I.e., p. 183; Carnoy, I.e., p. 268. 
 * Carnoy, 1. c, p. 271. 
 
24 
 
 Notes on the Syntax of the 
 
 turn for expressing the age of a person, which is rather unusnal 
 for Latin; only one example is found in these inscriptions: 
 2892, Ciii . . . an(nos) haben(ti) XXV. This is interesting 
 because of the similarity of the Romance idiom: French, j'ai 
 . . . ans ; Spanish, tengo . . . aiios. We may add to this 
 several examples in Gallic inscriptions and one in Petronius.^ 
 
 Here again separate consideration is desirable. Place where 
 is expressed both with and without the preposition in. In 
 pagan inscriptions the tendency is to adhere to the classical con- 
 struction and insert the preposition. Sufficient examples with- 
 out in, however, occur to show that the practice of omitting it 
 was well established among the people. In most cases, on the 
 other hand, the noun is qualified by an attribute and the irreg- 
 ularity, if it may be so called, is thereby diminished.- 4314, 
 Hoc . . . requiescunt ossa sepulchro; 2335, luvenis tumulo 
 qualis iacet; 3180, Qui militavit cohorte X; 3453, j^unc iacis 
 hoc tumulo. On the other hand the tendency to obliterate dis- 
 tinctions is shown by the use of the preposition with locus, which 
 is contrary to the spirit of classical Latin.^ The preposition is 
 used also when locus is qualified by an adjective. 1062, I(n) 
 loco quem, etc. ; 1064, 1066, In locum quem, etc. ; 3354, Iacet 
 ... in locum Campaniensem ; 3015, Con(c)uiescen(t)es in 
 locum. The preposition is omitted with the phrase suo solo, 
 which is, therefore, treated as the classical locust 1956, Porti- 
 ons ad balineum . . . solo suo . . . d. d. ; 1074, Porticus lapi- 
 deas . . . solo suo . . . d. ; 1459, Q. Memmius . . . statuas 
 duas . . . solo suo fecit. The same impossibility to generalize 
 is met with in Christian inscriptions, nor is it possible to draw 
 any definite line of distinction. The preference seems to t^nd 
 toward the omission of the preposition, but too many cases of 
 
 'C. I. L. XIT, 2141, 527G, Habere annos; Petron. 25, Puella quae non plus 
 quam septem annos habere videbatur. 
 = Pirson, I.e., p. 191. »G.-L., I.e., p. 247, n. 1. 
 
 * Draeger, I.e., 1. p. 525. 
 
 Latin Inscriptions found in Spain. 
 
 25 
 
 its use even with locus occur to make this statement strictly ac- 
 curate. I. H. C. 31, In hoc loco quiescentis; I. H. C. 483, In 
 ]ocu(m) hunc s(anct)um; I. H. C. 115. (I)n locum Xati- 
 vola ; I. H. C. 376, In f undum ; I. II. C. 174, In os tumulos ; 
 I. II. C. 33, In hoc saeculo . . . transegit; I. H. C. 57, In 
 hoc altario; I. II. C. 60, In hoc tumulo iacet; I. II. C. 189, 
 In isto loco sepultus est; I. H. C. 86, Celesti regnat in aula. 
 But: I. H. C. 533, Sepulta est tumulo; I. 11. C. 12, Hoc loco 
 . . . clegi quiescere proles. 
 
 In more than one case the preposition is used with the 
 proper name of a city, for the most part a late Latin usage, but 
 perhaps always more or less common in popular speech.^ 4587, 
 In Helviano . . .pos(i)tae sunt; I. H. C. 488, Kegnante 
 ... in Legione. In late Christian inscriptions the preposi- 
 tion ah may be used with the proper name of a city when 
 motion from it is expressed, also a vulgar Latin turn.^ I. H. C. 
 275, Yenit a Corduva. The preposition a or ah is usually in- 
 serted, as it is in classical Latin, with the name of a city, when 
 it is the point from which distance is measured. The viola- 
 tion of this principle,, however, is found in several cases, and 
 in most of them the name of the city begins with the vowel a, 
 so that the omission might be a graver's error. 4697, Abiano 
 ad oceanum; 4779, Aquis Flavis m. p., etc.; so 4780, 4781, 
 4782, 4783; 4872, 4870, Bracar Aug(usta) m. p. XIX; cf. 
 4803, a Bracar Aug(usta) ; so 4804, 4805, 4806. Verbs indi- 
 cating motion to a place are for the most part followed by the 
 Accusative with the preposition ad in regular fashion. Occa- 
 sionally such a verb is construed with m and the Ablative, so 
 that rest in a place is really the idea conveyed, though motion 
 to a place is meant ; even the Dative is found. ^ It will be noted 
 that these examples are found in very late inscriptions, in which 
 the difference between the cases is reduced to a minimum."^ 
 I. II. C. 244, Qui in hac aula . . . ingreditur; cf. 3871, In- 
 gressum militiam. 
 
 ^ Sclimalz, I.e., p. 255, par. 94 (end). 
 
 '^ G.-L., 1. c., par. 391, Rem. 1 ; Schmalz, 1. c., p. 254, par. 93. 
 
 ' Canioy, 1. c., p. 270. * Pirson, 1. e., p. 200. 
 
26 
 
 Notes on the Syntax of the 
 
 It remains to notice a few substitutions. Pirson ^ has shown 
 that the adjective urhicus, in Gallic inscriptions has come to be 
 the equivalent of Eomanus. In Spanish inscriptions the word 
 is used more freely still and is substituted for the prepositional 
 phrase ad urheni Romam or simply Romam. The reading seems 
 certain enough to justify this interpretation. 379, M. lul(io) 
 Serano in itinere ur(bico) defuncto et sepulto. In still another 
 case the Accusative of the name of a city is used instead of the 
 Locative, or apud with the Accusative. 4227, Decuriali allecto 
 
 Italicam. 
 
 Ablative Absolute. 
 
 This construction, always a favorite in Latin, is very fre- 
 quently used in Spanish inscriptions. In classical Latin, al- 
 though it might express a variety of ideas, it was hedged about 
 by definite restrictions and this very fact would expose it to 
 variations and incorrectness of usage in the mouths of the ig-no- 
 rant. In the Latin of Gregory of Tours, for example, the IvTomi- 
 native. Accusative and xVblative are all used in this construction 
 with practical identity of meaning.- In these inscriptions 
 abundant signs of the disintegration of the Ablative Absolute 
 appear. Of these one of the most noticeable is its replace- 
 ment by the Nominative Absolute in late inscriptions. 
 I. IL C. 275, Peracta sunt . . . Ordinius peragens sceptra; 
 I. H. C. 469, Peracta sunt . . . Garsea sceptra . . . pera- 
 gens. The best examples are those in which the Nominative 
 and Ablative are used in the same sentence, thus furnishing 
 additional evidence, were it needed, that the two construc- 
 tions were parallel at this late period.^ I. H. C. 472, Ke(ge) 
 Adefonso Suero Alvarie abas ; 3240, Factis pontibus et fistulis 
 et lacus. A good example of the misconception of the Ablative 
 Absolute and its consequent weakening, and at the same time one 
 that goes back farthest is the expression se vivo, a, is, and the 
 
 U.c, p. 267. 
 
 ' Carnoy, 1. c, p. 267. 
 
 ' Bonnet, 1. c, pp. 558 ff. 
 
 Latin Inscriptions found in Spain, 
 
 27 
 
 like, which seems to be common to all inscriptions.^ In a few 
 phrases and only a few is it correctly used. 3459, Viva sibi 
 fecit. It is possible that such examples contain the point of de- 
 parture. The next step may then have been something like 
 this: 4050, Se vivi fecerunt, the strict meaning of se having 
 faded out. Finally the still more impossible Ablative Absolute 
 makes its appearance. 2973, Atilia . . . se viva fecit; 3049, 
 Pomj)eia . . . se viva sibi . . . fecit; 3489, Numisia . . . se 
 viva fecit; 266, Flavins se vivo monumentum fecit; 3960, 
 Rufus coniimx se vivis comparai-nnt ; Eph. Ep. 8. 30, Me vivo 
 f. c. ; 507,, Ipsa sibi se viva . . . f. c. The climax of incor- 
 rectness is seen in a type like the following. 3509, Valeria . . . 
 se vivo fecit; 5721, (T)uravo . . . filio vivi fecit. 
 
 Interchange of Cases. 
 
 One of the most striking features of the language of inscrip- 
 tions, as indeed of late Latin in general, is the extent to which 
 confusion in cases appears and the ease with which one case is 
 substituted for another. Bonnet^ and Pirson^ have not only 
 discussed this point as it appears in the respective spheres with 
 which they are dealing, but have also assigned sufficiently 
 plausible reasons for it, the most potent of which are similarity 
 in sound of the various endings in rapid pronunciation and the 
 fall of m and s of the Accusative endings. In many cases, 
 nowever, minor reasons may be offered sufficient to account for 
 the substitution."* 
 
 Nominative and Genitive: — The peculiarities here may be 
 traced to habits of thought under the influence of which the 
 attention of the author was diverted from the strict syntactical 
 connection of words. For example, the names of persons, even 
 though in apposition with a word in some other case, were 
 likely to be felt as titles and written in the l^ominative. 6328^, 
 Manus Aureli Pacatiani filius possessor; I. H. C. 166, Hie sunt 
 
 ^Pirson; I.e., p. 185; Carnoy, 1. c, p. 268. 
 M.c, p. 522. M.c, p. 187. 
 
 * Carnoy, 1. c, p. 2C7. 
 
28 
 
 Notes on the Syntax of the 
 
 Latin Inscriptions found in Spain. 
 
 29 
 
 sepiilcra sanctorum . . . Xigrimis episc. Sefronius episc. ; 
 I.H.C. 228, Hie req(ui)escit corp(us) f (a)m(u)lorum dei 
 Agoblasto ( ?) Abeataresiis. Other examples, on the contrary, 
 seem to offer peculiarities for which no explanation and cer- 
 tainly no justification is at hand. I. II. C. 259, In nomine 
 dni . . . sive omnium eius sanctorum (?) ... Mariae virgi- 
 nis bissenisquc apostolis; I. II.C. 519, Kequiescit . . . Eulalia 
 mar(ty)ris. The influence of a contiguous word is seen in the 
 following: 5304, Luni Tapili f. . . . statuit ; I. II. C. 276, Hie 
 nos ea f(a)m(ul)a dei Eldonciae fundatum est. 
 
 Nominative and Accusative'. — By far the most interesting 
 examples of this type are those in which the Accusative takes 
 the place of a Predicate Nominative or even of the subject. 
 The examples are very late but are interesting as showing the 
 inclination for the Accusative as well as the extent to which 
 the cases were leveled. I. II. C. 494, Sit illi div(us) adiutorem 
 et protectorem; I. II. C. 135, Perfectum est templum hunc; 
 I. H. C. 396, Datum est . . . soledos III ; I. H. C. 474, Aulam 
 . . . fuit luteam. In the example 6180, Limenius, Euplium, 
 Patinicus, Calimorfus, Torax, ete. (cf. Hiibner's note on the 
 inscription), Euplium may be the pet name of a horse and so 
 put in the neuter. Here may be put the use of the Accusative 
 for the Nominative in the construction "" Attractio Inversa " ; 
 the examples seem worthy of recording.^ 4125, Sententiam 
 quam tulit L. Novius Kuf us ; L H. C. 474, Hanc quani cemis 
 aulam . . . fuit luteam. 
 
 Genitive and Vocative: — The Genitive is written for the 
 Vocative in one inscription by a misunderstanding of its form. 
 The sentence was begun in the form of direct address, which was 
 kept up until the word corpus was reached, where under the 
 influence of numerous inscriptions of a different type, the direct 
 form of the proposition was dropped and the Genitive ahhatis 
 instead of the Vocative ahhas, was attached as a limiting word 
 
 * Schmalz, I.e., p. 21G. par. 12; G.-L.. I.e., par. G17. 2. 
 
 to corpus, I. H. C. 142, Haec tenet (u)rna tu(u)m . . . 
 corpus Vincenti abb (at) is. 
 
 Genitive and Ablative: — Rather remarkable is the substitu- 
 tion in a single inscription of the Genitive for the Ablative, 
 which is more remarkable since it occurs in a well-understood 
 formula. 1459, Statuas duas . . . unam nominis sni alteram 
 patris poni iussit. The most plausible explanation here is that 
 the writer instead of following the usual phraseology suo . , . 
 et nomine patris has regarded the expression nominis sui as a 
 sort of substitution for a proper noun, and is dealing with it 
 just as with the phrase statuam Minervae. - In the following 
 epitaph the Genitive, though properly a limiting Genitive, may 
 be said to be a weak substitute for the classical Ablative of 
 Origin: I. H. C. 2, Sinticio . . . traens linea(m) Getarum 
 (= ducens originem ab Getis. Hiibner's note). 
 
 Accusative and Ablative: — The Accusative and Ablative are 
 easily confused with one another owing to loss of final m of the 
 foiTiier. This confusion occurs most readily after prepositions, 
 as Camoy's list ^ shows. The employing of the Accusative for 
 the Ablative is especially noteworthy in an inscription where it 
 clearly takes the place of the Ablative of Origin or Quality. 
 I. H. C. 372, Aurelius . . . nationem Afram qui vixit, etc. 
 
 As an appendage here may be put numerous examples of a 
 type that does not prove disintegration, nor throw light on 
 syntax, but that involves irregularities which arise simplv from 
 confusion in the mind of the graver as to the form of the in- 
 scription with which he was dealing; so that the proper noun 
 at the head of the inscription may be in one case and a word 
 in apposition further down in quite another. With these may 
 be compared Pirson's long list and discussion of same.^ 1099, 
 Maria Q. f. Nymphe M. Aureli uxori; 1262, M. Acenna . . . 
 quaestori . . . Illviro ; 2054, Imp. Caesari divi Nervae f . divo 
 Traiano optimo . . . pater patriae; 2126, C. Cornelius C. f. 
 . . . flamen Ilvir municipi pontifici, etc.; 2478, Imp. Caes. 
 
 * 1. c, p. 205 and p. 269. 
 
 •I.e., p. 166. 
 
30 
 
 Notes on the Syntax of the 
 
 Latin Inscriptions found in Spain. 
 
 31 
 
 Nerva Traiano Aug. Ger. Dacico; 4834, Imp. Caes. C. lulius 
 . . . et C. lulius Verus . . . principi iuventutis; 4858, Imp. 
 Caes(ar) C. lulius Vero Maximinus; 5293, Caecilia Q. f. 
 Tuscae ; 382, D. M. P. Lucani . . . Eomae sepulto. 
 
 Pronouns. 
 
 The accurate distinctions of classical Latin were not adapted 
 to the use of the people and many of them fall into disuse. It 
 would be safe to assume that the difference between suus and 
 eius or eorum would not be well preserved in popular speech and 
 Pirson ^ has shown by means of a long list of examples that the 
 boundary between these two in Gallic inscriptions is very nar- 
 row. His results apply equally well to Spanish inscriptions. 
 They show the crude way of thinking of some of these writers, 
 their careless method of forming sentences and their inability 
 to carry out an idea to its conclusion. These deficiencies are 
 well indicated, too, by the use of pronouns in the two inscrip- 
 tions following, in each of which the author misunderstood the 
 form of the inscription; in the one changing from first person 
 to third and in the second, from second person to third: 4323, 
 Aufidia Prima . . . sita sum posuit Florentinus eius (for 
 mens) ; I. H. C. 242, Is rex Alfonsi patris sui vestigio pru- 
 denter . . . regnum gubemas {sui for tui and is for tu, or else 
 guhemas for guhernat). 
 
 The curious appearance of what seems to be a syncopated 
 form of the demonstrative ille is worthy of note. The brief 
 document in which it is found is without date, but the reading 
 seems well attested and the example genuine. 3668, La tumba 
 Probae Publ. ingenuus ( ?) t. s. 1. There is good reason to be- 
 lieve that the initial syllable of ille was very lightly uttered when 
 in unemphatic position and appeared as probably (il)lum 
 patrem, (il)lam matrem, etc., in pronunciation, so that eventu- 
 ally the definite article developed out of these contracted forms. - 
 
 La, then, in the example given above may be taken as the 
 graver's effort to represent ilia as he heard it pronounced about 
 him, and may show that even at this period this syncopated 
 form was in vogue. ^ I^ote worthy is the for Latin redundant 
 use of the demonstrative pronoun as seen in a limited number 
 of cases. Many examples of this Graecism, if it may be here 
 so called, are given by Eonsch.^ Eph. Ep. 9. 176, Ut puellam 
 mitteres illam (cf. note on the inscription) ; I. H. C. 533, 
 Q(u)em . . . sepelivit eam. Another Graecism that occurs in 
 late and especially ecclesiastical Latin in which the relative 
 pronoun is regarded as a sort of definite article is apparently 
 found in one epitaph, unless one prefers to -regard uxori as in 
 apposition.3 5058, Cui benemeritae uxori . . . Chrysanthus 
 dignum honorem dedit. 
 
 ^ Meyer-Liibke, 1. c, 2, p. 135. 99 ff.; Rydberg, Geschichte des franzo- 
 sischen a, p. 271. 
 
 ^Itala und Vulgata, p. 444. 3. 
 ^ Ronsch, 1. c, p. 444. 3. 
 
 * 1. c, p. 203. 
 
 "Lindsay, The Latin Language, p. 167. 4. 
 
32 
 
 Notes on the Syntax of the 
 
 VERBS. 
 
 Tenses. 
 
 The use of the Present for the Future and vice versa is com- 
 mon to the language at all periods and for that matter to all 
 languages. Pirson ^ has outlined the extent of this usage in 
 the inscriptions of Gaul, and the statement there made will 
 apply also to Spanish inscriptions. The construction is essen- 
 tially vivid and colloquial. At any rate the two tenses seem 
 to be interchangeable at times and occasionally stand side by 
 side in the same sentence and construction, as indeed, else- 
 where in late Latin, so that it is impossible to tell which tense 
 was meant by the author to prevail. Uncertainty is further 
 caused by the easy substitution of e for i and vice versa. ^ 2262, 
 Tu qui stas et leges titulum meuni ; 5058, Dicite qui legetis, cf. 
 I. H. C. 232, Dicite qui legi(ti)s; 5241, Die rogo qui transis; 
 1235, Quisquis legis . . . sentis quam vixerim; I. H. C. 125, 
 Dum coelum et paradisum accipit homo; I. II. C. 142, Onmibus 
 his mox est de flammis flammas tollere. 
 
 When one action in the future occurs before another also in 
 the future, classical Latin required the antecedent action to be 
 written in the Future Perfect. The attempt is noticeable in 
 these inscriptions to observe this principle, but not infrequently 
 confusion is apparent and the Future Perfect is extended to 
 actions that are purely contemporaneous with the time of the 
 principal verb. It seems that the boundaries of the use of the 
 Future Perfect were made narrower, until after the best period 
 its substitution for the Future was confined almost entirely to 
 vulgar Latin, where apparently there was some inclination for it. 
 
 1 1. c, p. 206. 
 
 * Bonnet, 1. c, p. 634 ; Schmalz, 1. c, p. 334 ( top ) . 
 
 Latin Inscriptions found in Spain. 
 
 33 
 
 Its use for the Future in conversational idiom is familiar.^ Such 
 may be its use in the following examples, at any rate within 
 certain limits the two tenses appear to be interchangeable. 1276, 
 Quoat vixero . . . dividam; 1276, Quot praestabo dumtaxat 
 quoat vixero; I. H. C. 336, Siquis . . . inquietare voluerit sit 
 anathema. This halting between Future and Future Perfect is 
 further apparent in the formula die qui praeteris, etc. 1634, 
 Dicite qui legitis s. 1. 1. 1. ; 1126, Hie qui leg(i)s di(cas) ; 4174, 
 Bene sit viator qui me non praeteristi ; 1728, Te rogo praeteriens 
 cum legis ut dicas. But: I. H. C. 283, Obsecramus qui haec 
 legeritis ; I. H. C. 496, Depreco vos . . . qui ad hunc sepulcrum 
 intraveritis ; 1. H. C. 516, Qui legerit . . ..oret. In a single 
 inscription the inclusive formula in which the Present and 
 Future Perfect are used parallel with one another is employed 
 in imitation, perhaps, of well-known legal phraseology and is 
 due to a false striving after exactness. 172, Quos . . . Caesari 
 inimicos . . . esse cognovero . . . et siquis periculum infert 
 intuleritve si(quis) sciens fallo fefellero-ve, turn, etc. 
 
 The Perfect Infinitive as an object of a verb is uncommon 
 here as in the literature. Its use in a few cases is in conformity 
 with the construction found in the poets.^ 1414, Debueras . . . 
 habuisse mille (annos) ; 4314, Quibus nequeas opposuisse 
 manus. 
 
 Moods. 
 
 Infinitive : — Inasmuch as inscriptions contain very little nar- 
 rative and practically no lively nor spirited discourse, the His- 
 torical Infinitive rarely finds its way into them. One example 
 only of its use may be found in Spanish inscriptions and here 
 it is well adjusted to its surroundings. 6278 ^^, Statim sacer- 
 dotes . . . concursare gaudere inter se loqui. The Greek con- 
 struction of employing the Infinitive as a complement of ad- 
 jectives, so freely used by the poets and to some extent by later 
 prose writers also, is occasionally met with here. However, the 
 
 * Schmalz, I.e., p. 336, par. 217; Plautus, Cap. 194 (Lindsay's note) 
 »G.L., I.e., p. 180 (b). 
 
34 
 
 Notes on the Syntax of the 
 
 examples are drawn from poetry, so there is nothing incorrect 
 about the usage, which is by no means a favorite in inscriptions 
 of Spain. ^ 4314, l^ec tamen ignari flectere lora manu ; I. H. C. 
 361, Est homines vivificare potens. The same remarks apply 
 to the passive Infinitive with dignus in the case of which the 
 poet is acting within his rights in employing it in a metrical 
 inscription.^ I. H. C. 392, Dignus apostolico primus honore 
 coli. Several authentic cases of the Infinitive to express pur- 
 pose appear. It is essentially popular in sphere and is sup- 
 posed to have been adopted from vulgar Latin by poets and late 
 prose writers. All of the examples occurring in these inscrip- 
 tions are found in metrical epitaphs.^ 4427, Manes si saperent 
 miseram me abducerent coniugem vivere iam quo me lucem iam 
 nolo videre; I. PI. C. 158, Cum flamma vorax veniet comburere 
 terras. 
 
 Indicative in Indirect Questions: — The only peculiarity to 
 be noted in connection with the Indicative Mood is its use in 
 indirect questions. At bottom, it is only one of the many 
 evidences of the popular character of inscriptions, since this, like 
 the Indicative in deliberative questions, is a feature of colloquial 
 and poetical, as it is of late Latin usage. With these facts in 
 mind, it would virtually amount to a peculiarity were this use 
 of the Indicative not found in Spanish inscriptions, even though 
 it is not met with in classical prose."* 4426, Aspice quam 
 subito marcet quod floruit ante aspice quam subito quod stetit 
 ante cadit; 6130, Quare mortuus sum nescio; I. H. C. 219, 
 Quis quantusve fuit Samson . . . personat Esperio; I. H. C. 
 255, Omnis conventus conoscat q(u)orum . . . veneratur re- 
 liquias. 
 
 Subjunctive for Imperative: — In certain legal documents 
 
 1 Kiihner, I.e., 2, p. 504 (d) ; Draeger, I.e., 2, p. 378. 3. 
 
 'Draeger, I.e., 2, p. 331, par. 421. 
 
 «Draeger, I.e., 2, p. 367, par. 433; Kuhner, I.e., 2, p. 501, 6 (a); 
 Honscli, 1. e., p. 447. 
 
 *Draeger, I.e., 2, p. 474; Ktihner, I.e., 2, p. 989. 2 with referenees; 
 Eiemann, Etudes sur la langue et la grammaire de Tite-Live, p. 301, n. 
 3 and p. 10, n. 1; cf. Prop. 3. 5. 26flf.; Bonnet, 1. c., p. 676 and refs. 
 
 Latin Inscriptions found in Spain, 
 
 35 
 
 found in these inscriptions notably 5181, 5439, 1963, 1964, the 
 Second Imperative is almost exclusively used and this is in con- 
 formity with Latin usage in such documents. This very fact 
 makes the one example of the Subjunctive found in an inscrip- 
 tion like the following worthy of notice:^ 5181^^, Profite- 
 antur et salvant (cf. note). 
 
 Mood after Conjunctions: — Little out of the ordinary is 
 found, but we may discuss the following: Peculiar is the em- 
 ployment of dum with the Subjunctive in a manner exactly 
 parallel with the use of cum, however, retaining its distinctive 
 meaning of " while, as long as, etc." This -may be due to the 
 fact that dum = while and cum = when are separated by very 
 narrow limits, and to a consequent blending or obliteration of 
 the limits of the two. The construction is also found in the 
 Latin of Gregory of Tours.^ I. H. C. 12, Dum simul . . . cum 
 viro carpere(m) vitam ... me fortuna tulit; I. H. C. 65, 
 lamq(ue) novem lustris gaudens du(m) vita maneret. In still 
 another example dum has both the meaning and construction of 
 cum,^ I. H. C. 385, Dumque supremus litui crepor concusserit 
 orbem. In dextera . . . queat tunc sorte(m) adesse. Forsi- 
 tan appears probably only once in all Spanish inscriptions and 
 their tendency to depart from classical models is shown in this 
 instance by the use of the Indicative; unless one prefers to 
 regard it as an adverb here rather than as a conjunction as 
 sometimes in Livy.^ 4314, Favisti vivo forsitam ipse mihi. 
 In apparently the only case in which quamquam occurs, it is 
 found in the Subjunctive. 6278 ^'^, Quamquam . . . nonnulli 
 arbitrentur. 
 
 Impeesoi^al Veebs. 
 
 Such Impersonal Verbs as pudet, piget, were used personally 
 in the literature, but almost wholly in early or late Latin. The 
 examples that can be cited, though not numerous, seem to show 
 
 * G-L., 1. c., par. 268. 2. 
 
 * Bonnet, I.e., p. 319; Schmalz, I.e., par. 306. 2. 
 
 ^ Kuhner, I. e., 2, p. 909, Rem. 3. * Riemann, 1. c., p. 292. 
 
36 
 
 Notes on the Syntax of the 
 
 this.i An example of each of the above verbs used personally 
 can be given from Spanish inscriptions and it is possible that 
 here, too, the influence of misereor is at work. I. H. C. 496, 
 Depreco vos . . . pro me Adefonso orare non pigeatis; 4284, 
 Si sordes, patior ; sed pudet hospitium. 
 
 Regimen of Vekbs. 
 
 In the matter of construction, verbs as employed in Spanish 
 inscriptions closely resemble the verb of Silver Latin. They 
 represent a wide sphere of extension over classical usage, as 
 judged from the government of those with which we meet. The 
 following list presents peculiarity either by wholly departing 
 from the classical norm or by adopting a regimen not often 
 used in the best period. They may be grouped broadly under 
 
 two heads : 
 
 1. Verbs that rarely or never take an Infinitive construc- 
 tion in good Latin are loosely so employed. 
 
 Cavere, — In the example to be cited cavere seems to be 
 identical in meaning with curare. The occurrences of cavere 
 with an Infinitive are not many at any period of the language ; 
 the examples given by Draeger - with few exceptions are all 
 equivalents of the Imperative, so that this example appears to 
 be rather out of the ordinary for prose. 1359, Qui . . . caverat 
 sepulcrum sibi fieri. 
 
 Concedere. — The Infinitive construction with this verb is 
 usual only when the passive is used. In the active it is rare 
 even in the poets.^ 89, Quai fate concesserunt vivere. 
 
 Decernere. — Here the Infinitive construction is evidently by 
 analogy to inhere, cf. 964, Statuam poni iussit. ISTumerous 
 examples of this kind might be given. Draeger ^ cites a single 
 example similar to that given below and this is taken from an 
 African inscription. Such an example as: 1130, Statuam 
 
 1 Neue, 1. c, 3, p. 653. 
 
 = 1. c, 2, p. 334; G-L., 1. c, par. 548. 3. 
 
 " Draeger, 1. c, 2, p. 330. 
 
 * 1. c, 2, p. 407. 
 
 Latin Inscriptions found in Spain. 
 
 37 
 
 ponendam decrevit gives the normal construction in these in- 
 scriptions after decernere. 5522, Statuam equestrem poni de- 
 crevit Quintia. 
 
 Eligere. — I. H. C. 12, Hoc loco erga meos elegi quiescere 
 proles. 
 
 Mereri. — With an Infinitive first in Propertius.^ I. H. C. 
 96, Mereamur ingredi paradisi i(a)nue. 
 
 Permittere. — The Infinitive here belongs to the sphere of 
 Silver Latinity.^ I. H. C. 501, Ut non permitas introire 
 angelum, etc. 
 
 Praecipere. — In late Latin praecipere became equivalent to 
 inhere and is often so employed by the Church Fathers.^ 4853, 
 (Y)ias et pontes . . . restitui praec(e)perun(t). 
 
 Sancire. — The customary regimen of sancire when followed 
 by a clause is ut -j- Subjunctive. Draeger ^ mentions the In- 
 finitive construction, but is not definite. Cicero lexicons reveal 
 no example similar to those appended. I. H. C. 253, Hanc 
 aulam construere sancserunt; I. H. C. 259, Hanc aedificari 
 sanxit munitionem. 
 
 2. Verbs of miscellaneous constructions. 
 
 Credere. — This verb in a late Christian inscription shows the 
 effect of the late Latin extension of the quod construction, em- 
 ploying it instead of the Accusative -j- Infinitive.^ I. H. C. 95, 
 Credo quod redemptor mens vivet. 
 
 Curare. — In a single inscription curare is followed by the 
 Past Participle instead of the Gerundive which is its ordinary 
 construction. It is a usage peculiar to early Latin ; only missam 
 facere occurs in the classical period.^ 453, Factum curavit 
 
 * Draeger, I.e., 2, p. 331; Pirson, I.e., p. 214. 
 
 * Draeger, 1. c., 2, p. 330; Kiihner, 1. c., 2, p. 809.3. 
 
 •Schmalz, I.e., p. 291. M.c., 2, p. 249. 
 
 " Ronsch, 1. c, p. 427 ; Mayen, De Particulis quod quia quoniam quo- 
 modo ut pro Ace. cum Infinitive post Verba sentiendi et declarandi positis. 
 Diss., Kiel, 1889; Plautus, As. 52 (Gray's note). 
 
 •G. L., 1. c., par. 537.2; Kahner, 1. c., 2, p. 572 (d). 
 
38 
 
 Notes on the Syntax of the 
 
 Progela, etc. In still another example it is found with the 
 Infinitive, which again is limited in usage.^ 4788, Restituere 
 
 (c)urarunt. 
 Dicere.— Somewhat similar to the regimen of credere ]ust 
 
 discussed is that of dicere in an epitaph where it is followed by 
 
 an ut clause. It is barely possible to regard it as introducing 
 
 an Indirect Question.^ I. H. C. 232, D(i)cete qui leg(it)is ut 
 
 me peccat(o)rem memor habeas ( ?) sacra ex oratione. 
 
 luhere, This verb to all appearances presents the unique 
 
 construction of the Infinitive combined with a Subjunctive 
 clause in the same sentence. Somewhat similar instances are 
 seen in the Latin of Gregory of Tours.^ I. H. C. 494, Iuss(i)t 
 renovari . . . et pro tali laborem sit illi . . . adiutorem. 
 
 Nolle,— The Ace. + Inf. with same subject is not usual with 
 this verb.^ 4427, Me lucem iam nolo videre. 
 
 It remains to call attention to the frequency with which ut 
 is omitted after verbs of asking, beseeching, etc., so charac- 
 teristic of Petronius and of conversational idiom in general.^ 
 462, Te rogo oro obsecro uti vindices; 558, Oro ut . . . dicas. 
 But: 540, 958, Te rogo . . . dicas; 1235, Hoc pe to nunc dicas; 
 1580, Te lapis obtestor . . . residas; I. H. C. 242, Quaesumus 
 . . . dei gratia parcat ei. 
 
 Compound Verbs. 
 
 Here may be treated the use of the simple for the compound 
 verb and vice versa. It has already been emphasized by those 
 who have touched upon this subject that the use of the simple 
 for the corresponding compound verb and vice versa is a feature 
 of poetical diction and, with limitations, of certain prose writers 
 
 *G-L., I.e., par. 423, n. 2; Kiihner, I.e., 2, p. 491 (e) ; Pirson, I.e., p. 216. 
 * Ronsch, 1. c., pp. 445 and 427. 
 
 •Bonnet, I.e., pp. 667 and 668; Saflfold, The Constructions with lubeo. 
 Johns Hopkins University Diss., 1902. 
 *Draeger, I.e., 2, p. 402 (c) ; Schmalz, I.e., p. 290, par. 154. 
 •G-L., I.e., par. 546.2, Rem. 2. 
 
 Latin Inscriptions found in Spain, 
 
 39 
 
 of the Augustan Age and after. In both it may be due largely 
 to striving after rhetorical effect. It is also a characteristic of 
 the sermo pleheius and one would expect to find something of this 
 interchange in inscriptions, which are fairly representative of 
 this type.^ If the subject be fraught with difficulties in dealing 
 with it as it appears in the literature, it is none the less so in a 
 field like epigraphy. However, some evidence of this phenome- 
 non seems to appear in these inscriptions, and, though it is im- 
 possible to be altogether definite in view of the small amount of 
 material available and the consequent impossibility to duplicate 
 examples, yet a number of these cases may be worth noticing 
 and as far as practicable strengthening with citations from other 
 
 sources. 
 
 Cludo for Includo,^ — 4550, Cuius basis lapidea acre clusa; 
 1821, Herenia . . . inclusa hoc tumulo; I. H. C. 144, Inclusi 
 tenerum praetioso marmore corpus. 
 
 Efficio for Facio. — I. H. C. 245, Pontifex effectus. 
 
 Fero for Aufero or other compounds.^ — 1413, Viginti . . . 
 tecum nam fers . . . annos; I. H. C. 34% Te matremque tem- 
 pus enorme tulit; I. H. C. 12, Me fortuna tulit; I. H. C. 223, 
 Ad eroas latus (e)st; cf. 1414, LXX tecum transfers non 
 amplius annos; 6064, Cui Parca vitam abstulerit praemature. 
 
 Fungor for Defungor. — I. H. C. 217, Functus in hac aula 
 quiescit. Examples of defungor in this sense are plentiful. The 
 reverse with a different meaning for the verbs may occur: 4261, 
 Defuncto honoribus. Here functus is practically always found, 
 occasionally perfunctus. 4526, Omnibus honoribus in rep(ub.) 
 sua perfuncto; 4205, 4201, Ob legationem qua . . . funct(us) 
 est; 4203, 4207, Omnib(us) honorib(us) . . . functo. 
 
 Infero for Aufero. — I. H. C. 249^, Quisquis inferre . . . 
 donaria nostra presumpserit ; cf. I. H. C. 256, Quisquis 
 auferre . . . donaria nostra sumpserit. 
 
 1 Wilson, Transactions A. P. A. XXXI, 1900, p. 205 and refs. 
 ^Tac., Ger. 45. Animalia . . . quae implicata humore mox durescente 
 materia cluduntur (Furneaux' note). 
 'Riemann, I.e., p. 194. 
 
40 
 
 Notes on tie Syntax of the 
 
 Loquor for Conloquor.—mS l^ Inter se loqui. 
 
 Pleo for Impleo.'—Q27S^% Tantisper plendi numeri gre- 
 
 ffariorum gratia. 
 
 8cando for Conscendo.^-l. H. C. 255, Scendens Xrs m 
 altum. This is frequent enough in good prose, but as a rule 
 conscendo is used in these inscriptions. 
 
 Teneo for Contineo or other compounds,^— 1. H. C. 142, Ilaec 
 tenet (u)ma . . . corpus, etc. cf. I. H. C. 363, Ketinet quos 
 Corduba; I. H. C. 129, Ista vorax fossa Dommici contmet ossa ; 
 I. H. C. 123, Haec cava saxa contine(n)t membra. 
 
 ' Transeo for Praetereo^—^lSl, Qui transis lege. In the 
 literature these two verbs appear to be interchangeable, but a 
 casual glance at Spanish inscriptions will show that in this 
 formula praetereo is practically always employed. The above 
 example is the only one of transeo. 
 
 To these may be added certain others of frequent occurrence, 
 where the use alternates between simple and compound and 
 which may be said to be practically interchangeable everywhere. 
 It is superfluous to name them at any length. Such are : decoro 
 and conc^ecoro ; Eph. Ep. 8. 194, Monumento hoc condecoravit : 
 condo and recondo; I. H. C. 85, Hie sunt reliquiae conditae; 
 I. H. C. 88, Hie sunt reconditae: iungo and coniungo; I. H. C. 
 34% Tumulo iungeris ; I. H. C. 213, Atletis iungitur rite; I. H. 
 C. 255, His pariter co(n)iunctae sunt; I. H. C. 218**, Coniunc- 
 tus mansit: curro and percurro; 5975, Quisquis in has partes 
 quisquis percurr(is) : moveo and promoveo; 6278^7, Lanistas 
 etiam promovendos vili studio: sacro and consecro; 2243, C. 
 Valerius ... hie consecravit; I. H. C. 172 and 175, Ecclesiae 
 sacratae sunt; I. H. C. 100.4, Consecrata e(st) baselica haec. 
 
 Prepositions. 
 
 ^ or Ah.— The only peculiarity here is one of form. In sev- 
 eral inscriptions, a is used instead of ah before a word beginning 
 
 » Festus, p. 291, de Ponor. «Riemaim, I.e., p. 198. 
 
 * Riemann, 1. c, p. 197. 
 
 *Cf. Vergil, Aen. 5. 326 ( Conington's note); Prop. 3. 7. 27. 
 
 Latin Inscriptions found in Spain. 
 
 41 
 
 with a vowel. Other examples do not appear to be at hand.^ 
 4895, 4896, 4899, A Augustobriga. 
 
 Cum. — The analytical tendency of this period of Latin is 
 showm by a late Christian inscription in which means is ex- 
 pressed by the aid of cum. This may be styled an ante-classical 
 and post-classical usage. Many examples occur in the Latin of 
 Gregory of Tours.^ I. H. C. 238, Cum dei adminiclo Erankila 
 abba condidit opus. The Ablative, usually with the preposition 
 cum, is regularly employed as in classical Latin to express the 
 appurtenances of a thing, in particular of grounds and building's 
 dedicated to the gods. To this, one possible exception is noted in 
 Spanish inscriptions and here the simple Ablative is used. The 
 text, however, is probably corrupt. 3960, Compararunt locum 
 uti est concameratum parietib(us) balneis hortis monumentum 
 construxerunt. cf. 1163, Sig(num) (aere)um . . . cum base 
 d.d. ; 1956, Porticus ad balineum solo suo cum piscina et signo 
 Cupidinis . . . d.d.; 2326, Yenerem Aug. cum parergo; 3240, 
 Factis pontibus et fistulis . . . cum suis omamentis. 
 
 j)e, — The Ablative with de is used to indicate the means by 
 which something is erected and is especially common in the 
 phrases de mo, de meo, etc.^ It has been shown by Pirson ^ that 
 in Gallic inscriptions means is seldom expressed by the simple 
 Ablative as in fecit impensa sua. In Spanish inscriptions, how- 
 ever, the simple Ablative in such phrases is found almost, if not 
 quite, as frequently as the prepositional form de sua pecunia. 
 The preposition is exclusively used only in de suo posuit. 1277, 
 1341, 2035, 2041, Sua pecunia posuit; 1250, 1367, Impensa sua 
 fecit; 1724, Impensa sua extruxit; 2070, Sumptu publico 
 posuit; 3570, Sua pecunia restituerunt ; 3240, 3280, 5181 ^^ 
 Sua omni impensa. 
 
 Many cases of de and the Ablative instead of the Partitive 
 Genitive occur in examples where classical Latin furnishes a 
 
 * Neue, 1. c., 2, pp. 827 ff. 
 
 'Bonnet, I.e., p. 603; KUhner, I.e., 2, par. 93.3 (c) 7; Schmalz, I.e., p. 
 272, par. 128. 
 
 ■ Draeger, 1. c, 1, p. 628. * 1. c., p. 197. 
 
42 
 
 Notes on the Syntax of the 
 
 sufficient parallel. Eph. Ep. 8.128, Aemilio Balaeso . . . cog- 
 natio de ceii(turia) ; 1585, M. Pompeius . . . primus de familia 
 Pompeia; Eph. Ep. 8.30, Si qui de meis. Also in places where 
 no pronoun or word upon which the expression depends is 
 written: 5439 ^-^-^^^ Si quis accusator de suo tempore alteri con- 
 cesserit. But the frequency of this partitive use of de, as well 
 as its tendency to supplant the Genitive, is best illustrated in 
 late Christian inscriptions, where the best defined examples may 
 be seen.^ These examples mark an advance over the same or 
 similar construction found in classical Latin in that here the de 
 phrase and the Genitive are not only parallel, but practically 
 equivalent, as shown by their use in the same sentence. I. H. C. 
 175**, Recondite sunt ic reli(quiae) cruore . . . pane . . . 
 cruce . . . de sepulcro . . . crucis ; I. H. C. 255, Intra prae- 
 ciosisima praesentis arce lat^ra hoc est de ligno plurimum sive 
 de cruce d(omi)ni; de vestimento illius . . . de pane . . . de 
 sepul(c)ro . . . atque sudario et cruore . . . de vestimentis 
 . . . de lacte; quaedam . . . et reliquiae . . . ho(c est) . . . 
 de S(an)c(t)o Tome, s(an)c(t)i Bartolom(ei), (d)e omnibus 
 ap(osto)lis et de aliis quam plurimis s(an)c(t)is; I. H. C. 263, 
 Hie sunt reliquie S. Mariae de lacte eius, de S. Paulo, de S. 
 Vincent., etc. ; I. H. C. 471, Hie siAut reliquiae reconditae, id 
 
 s(unt)t d(e) cruore dni de ligno dni de sepulcro . . . See 
 Marie, s(an)c(t)or(um) Petri et Pauli; I. H. C. 260 (see p. 
 120 addenda), Reliquiae que ibi sunt: de ligno dni, . . . de 
 pane dni, de sepulcro dni. By the side of these we may put 
 certain cases of the real Partitive construction as it exists in Old 
 Spanish, where it occurs with no great degree of frequency.^ 
 Poema del Cid 481, Tanto traen las grandes ganancias, muchos 
 gailados de averes e de vacas e de ropas e de otras riquiezas; 
 Poema del Cid 673, ^o nos daran del pan ; Libro de Alexandre 
 770^, Priso luego la manga e saco de los granos; Libro de 
 Apolonio 289*^, Priso del puro balsamo. This analytical ten- 
 
 ^ Juvenal, ed. H. L. Wilson, Introd., p. xlvii (d) and references; Reisig, 
 Lateinische Spraehwissensehaft, 3, p. 694 (note). 
 * Meyer-Lubke, 1. c, 3, p. 404, par. 366. 
 
 Latin Inscription found in Spain. 
 
 43 
 
 dency is pronounced in a few examples where de + Ablative 
 takes the place of the simplest form of the Latin Genitive. ^ 
 I. H. C. 472, Episcopus de Sanctae Mariae feci ; I. H. C. 284, 
 Rogare pro anima de d(omini)co Belazani; I. H. C. 34, Ego 
 Simeon filius de Rebbi, etc. 
 
 Ex, Apparently ex gives a loose way of expressing means in 
 
 an inscription too corrupt to deal with accurately (cf. note on 
 the line). I. H. C. 276, Mee arcitecton fuit facta ex mea 
 munera emta hec domus. In Cicero the verb recreare regularly 
 takes the preposition ex with the noun indicating that from 
 which one is recovering, while the simple Ablative marks the 
 means. From Spanish inscriptions we may cite the following : 
 2036, Ancipiti morbo recreatus. 
 
 7^._With the signification of " in accordance with," " with 
 respect to,'' the preposition in goes back to the early period of 
 the language, but is limited to a given number of formulae. 
 Cicero has the expressions in formulam, in sententiam; Plautus, 
 in leges, in formam, in faciem, in speciem, etc. Naturally one 
 finds the list of such examples lengthened in later writers. 
 The expression in verba seems not to appear prior to Livy and 
 Quintilian. It is in conformity with the usage of this school 
 that we find it in these inscriptions.^ 5406, ISTeve ut in ea 
 verba quae in verba . . . dari solet repromitteret. 
 
 With verbs of motion at this period there was a wavering be- 
 tween the simple Accusative and an Accusative construction 
 with the appropriate preposition. (See the construction with 
 ingredior, p. 25). It is doubtless due to this uncertainty that 
 the preposition in is written after one of these verbs, where in 
 good Latin a simple Accusative of the Direct Object is usual and 
 sufiicient.3 4514, Consecutus in honores aedilicios. 
 
 Followed by the Ablative in is used in a few examples ap- 
 parently expressing means.^ I. H. C. 143, Posside constructam 
 
 » Bonnet, 1. c, p. 611 ; Pirson, 1. c, p. 196 and ref. 
 « Draeger, 1. c, 1, p. 656. ' Carnoy, 1. c, p. 267. 
 
 *Draeger, I.e., 1, p. 652.12; Sehmalz, I.e., p. 275; Kuhner, I.e., 2, p. 
 408. 3. 
 
44 
 
 Noies on the Syntax of the 
 
 \mm\ 
 
 'ill' 
 
 ii 
 
 in aeterao munere sedem ; I. H. C. 363, Factis celebrantur in 
 istis. In others it occurs in a way almost parallel with cum, 
 perhaps traceable to the influence of the Vulgate.^ I. H. C. 33, 
 Eequievit in pace; I. H. C. 335, Sepelivit earn hac tumulo 
 ecclesiae in pace; cf. I. H. C. 186, Vixit . . . cum pace; I. H. 
 C. 34, Ingrede cum pace. It is further seen in a certain number 
 of metaphorical relations for which parallels may be found in 
 the literature.2 I. H. C. 413^, Praefluus in lacrimis ; I. H. C. 
 277, Momus in corpore vivens fecit; 4967.40 (a). Vivas in 
 Christo, (b) Spes in deo. 
 
 In the examples just considered the preposition is analogous 
 to the English idiom in similar cases. In one or two more it is 
 not only parallel with English, but at the same time further 
 removed from standard Latin usage. I. H. C. 215, Annos in 
 tempore quattuor; I. H. C. 396, Datum est . . . in auro soledos 
 III. cf. Spanish pagar in auko.^ 
 
 It is customary in Spanish inscriptions, as in others also, and 
 in" classical Latin for the word nomen to be used in the simple 
 Ablative to express in whose name a given action is performed 
 or anything is donated. In a few late inscriptions in is found 
 inserted with nomen and may possibly be explained by the 
 growing tendency to resort to the analytical construction, as in 
 the case of de already noticed, and so to simplify the complexity 
 of classical Latin. I. H. C. 259, 265, 267, 269. In nomine 
 domini dei, etc. Deserving mention only for its strangeness is 
 an example of in like the following, where Hiibner attempts to 
 explain its presence by confusion on the part of the author with 
 intemperies: 4853, (Via)s et pont(e)s in temporibus vetustate 
 conlap(s)os. 
 
 luxta. — The use of iuxta with the meaning of " according 
 to " is distinctlv late.** 191, Thermae . . . renovatae a solo 
 iuxta iussionemVumeri, etc. 
 
 * Pirson, 1. c, p. 198. * Draeger, 1. c, 1, p. 652. 10. 
 
 « Meyer-Liibke, 1. c, 3, p. 526. 
 
 •Draeger, I.e., 1, p. 587.3 (b) ; Hand, Tursellinus, 3, p. 542.7; Pireon, 
 I.e., p. 273. 
 
 Latin Inscriptions found in Spain, 
 
 45 
 
 
 Per. — The ordinary prose construction used in referring to 
 the building or location of a bridge upon a river was in + 
 Ablative. In the example below, per is employed and the ex- 
 planation probably is because via is metaphorically used for 
 pons, so that per applies to the former, in to the latter.^ I. H. 
 C. 23% Perdiderat usum suspensa via per amnem. 
 
 Pro, — Pro is employed only in late inscriptions with the 
 object for which something is given in exchange and is so con- 
 strued in the literature also.^ I. H. C. 494, Aveat pro tali 
 facto . . . remuneracionem ; I. H. C. 396, Datum est pro locello 
 . . . soledos III. In classical Latin, pro was attached to cer- 
 tain verbs, such as pet ere, supplicare, etc., with the appropriate 
 idea of " in behalf of." In later writers it was extended to other 
 verbs, such as timere, sollicitari, etc.^ It is by analogy with 
 these and especially with orare that in these inscriptions pro 
 is found with rogare, which at this period had taken on the 
 meaning of " pray " in an intransitive sense. I. H. C. 284, 
 Rogare pro anima, etc. 
 
 Secundum. — This preposition deserves mention only on ac- 
 count of the unusual phrase of which it forms a part. Here it 
 means " along,'' as in Latin of all periods.^ But it appears in 
 the phrase secundum parietem as a substitute for in fronte, a 
 turn probably occurring nowhere else (cf. editor's note). Eph. 
 Ep. 9.78, Primigenia . . . h. s. e. s. t. t. 1. ; secundo pariete 
 p. VII. In a. p. X. 
 
 Sub. — Suh is regularly employed in set phrases to denote 
 time.^ L H. C. 328, Sub mence i^ovembres; L H. C. 338, 379, 
 304, Recessit in pace sub die idus, etc. It is further employed 
 in a variety of phrases where it was but sparingly used by clas- 
 sical ^vriters, but more frequently by later authors. 518, Sub 
 cura; 1423, Sub nomine meo; I. H. C. 472, Sub Chr(ist)i nme; 
 L H. C. 244, Sub Christi . . . gratia . . . fecit; L H. C. 361, 
 
 1 Draeger, I.e., 1, p. 647 (d) ; Kiihner, I.e., 2, p. 408 (top). 
 
 2 Draeger, I.e., 1, p. 643.3. 'Draeger, I.e., 1, p. 643.2. 
 * Draeger, I.e., 1, p. 600 (b). 
 
 'Draeger, I.e., 1, pp. 661-662; Schmalz, I.e., p. 276, par. 137. 
 
46 
 
 Notes on the Syntax of the 
 
 Latin Inscriptions found in Spain. 
 
 47 
 
 Sub terrore . . . metuentes ite; LH.C. 470, Sub virtus 
 Christi; LH.C. 470, Sub misericordia et gratia dommi. 
 Separate citation is merited by the phrase: 4514, Sub eadem 
 forma. It is apparently an obvious parallel to the phrases sub 
 formula, sub lege of classical authors.^ In certain other connec- 
 tions this preposition is superfluous, since its omission would 
 constitute an Ablative Absolute construction. However, similar 
 examples are found in Livy.^ L H. C. 519, Sub Daciano pre- 
 side; I. H. C. 469, Sub valente . . . Adefonso prmcipe 
 
 ;Si/per.— Occasionally super is used to denote excess. 1. 11. O. 
 470 Era LXXVIII super millesima. But especially is it 
 allied with later usage in the phrase mper omnes.^ 4105, Su- 
 (per) omnes retro princi(pes) providentissimo ; 4112, Su(per) 
 omnes reliqu(os) praesedes iustissimo; I. H. C. 362, Super 
 omnia castos. 
 
 Adverbs. 
 
 From Demonstrative and Kelative Pronouns are formed a set 
 of adverbs indicating place and direction which, in standard 
 prose, are marked by the frequency as well as by the carefulness 
 and accuracy of their use. The ' whither,' ' where ' and 
 ' whence ' adverbs were restricted each to its proper sphere 
 and no infringement of one upon the territory of the other 
 was allowed. Just as in English ' there ' and ' where ' ^ are 
 used indifferently of situation and motion to a place, so it is 
 probable that in colloquial Latin this same confusion was usual 
 and accepted.^ The loss of distinction may also be inferred 
 from the Komance derivatives.^ In Spanish inscriptions and 
 especially in the later ones it is just this imperfect distinction 
 between these classes of adverbs that is noticeable ; so that the 
 boundary between hinc and hie, ibi and eo becomes extremely 
 
 ^Drager, I.e., 1, p. 661 (e) and (f) ; Kiiliner, I.e., 2, p. 302, Rem. 31. 
 
 ^Draeger, I.e., 1, p. 661 (e) ; Ktihner, I.e., 2, p. 415.4. 
 
 ^Draeger, I.e., 1, p. 663.3; Schmalz, I.e., p. 276.138. 
 
 *Ap., Met. 9.39: Ubi duels asiniim istum; Hand, I.e., 3, p. 165.9. 
 
 » Meyer-Lubke, 1. e., 3, pp. 531 ff.; cf. p. 486. 
 
 faint, if not quite obliterated. Just as in the case of hoc and 
 hunc, similarity of sound in the pronunciation of hie, hue and 
 hinc would be sufficient to account for their loss of identity, 
 and analogy to these might have influence in the case of ibi, 
 ubi and eo. 2592*^, Hinc sepulta est; 4386, Lolia Porci f. 
 hinc (s)ita; 4402, Kubena . . . hinc sitast; I. H. C. 86, Hie 
 sursum rapta; I. H. C. 110, Hu(c con)ditae sunt; I. H. C. 
 474, Adduxerunt ibi de urbe Avila. 
 
 In Silver Latin, as well as in the language of a later period, 
 appears quite a considerable class of adverbs formed from the 
 prepositions ex and de combined with an adjective. Such are: 
 ex adverso, ex diverso ^ and the Ovidian e duro, e facili, ex 
 aequo.^ Here would seem to be indicated growth in the forma- 
 tion of such adverbial phrases. To these, it appears, may be 
 added the rare e recenti, which, though analogous to those just 
 mentioned, is apparently found only in the passage given below : 
 5181 ^ (Aena quibus utetur) lavare tergere unguereque adipe 
 e recenti. IS'ot only was the list being extended in this direction, 
 but new and stronger adverbs were being formed by combining 
 a preposition with a simple adverb, a common expedient in 
 classical Latin, where it was limited to a few forms like exinde, 
 exadversum, subinde, de super and the like found at all periods 
 of the language. In Silver Latinity there is a noticeable in- 
 crease in these compounds. We may cite circumundique from 
 Statins and Gellius, circumsecus from Apuleius, and others.^ 
 We may further add ex tuncj^ ex invicem,^ abistinc,^ exsuperJ 
 It is evident, then, that exundique is due to the increasing incli- 
 nation for such compounds in late Latin and that the preposition 
 is merely added for emphasis, because the simple adverb was felt 
 to have lost directive force. I. H. C. 469, Hoc templum deco- 
 rum . . . exundique amplificatum erigitur. 
 
 « Madvig, Adv. Crit. 2, p. 80. 
 * Ronseh, 1. e., p. 233. 
 
 *Draeger, I.e., 1, p. 633 (e). 
 ' Draeger, 1. e., 1, p. 126. 
 « Souter, Arehiv 10. 412. 
 
 ' Johnston, The Querolus, p. 58. Johns Hopkins University Diss., Toronto, 
 1900. 
 ^ Carnoy, 1. e., p. 266. 
 
48 
 
 Notes on the Syntax of the 
 
 The adverb usque in classical Latin was employed m a variety 
 of connections, chiefly, however, in strengthening prepositions 
 and adverbs of time or their equivalents. This process of at- 
 taching it to various adverbs widened until in post-classical times 
 it might be used with practically any adverb or preposition. 
 So we find in Spanish inscriptions, I. H. C. 259, Usque nunc, 
 said by Draeger ^ to be a Vitruvian phrase. 
 
 The deviations from normal usage in Spanish inscriptions are 
 less frequent than in the inscriptions of Gaul. No such long 
 lists of peculiar constructions, of new and unfamiliar words 
 reward the seeker here as in Gallic inscriptions, nor is the exten- 
 sion in every field quite so wide in the former as m the latter. 
 This is due in part to the fact that in Gallic inscriptions the 
 body of material is larger, in part it may be due to the thorough- 
 ness with which Spain as a province was Eomanized. Never- 
 theless, the evidence of their popular character is patent and 
 complete, and harmonizes thoroughly with the facts hitherto 
 ascertained in regard to Folk Latin. Further it is largely on 
 the basis of the Christian inscriptions that such a conclusion 
 may be reached; often one must turn to these to find examples 
 of that which varies widely from classical usage. If it be pos- 
 sible to classify these inscriptions even in the most general way, 
 it mav be said that in syntax, vocabulary and style they seem 
 to bo "^allied with Silver Latin. Deviations from the usage of 
 Cicero are numerous, but they usually take us into the domain 
 of Silver Latinity. Not often does a usage occur in the better 
 class of documents that cannot be duplicated in that sphere or 
 actually traced back to it. Further than this, the influence of 
 the Augustan poets is unmistakable, chiefly in the metrical 
 epitaphs, where we find manifest imitations of them, not only 
 in the extension characteristic of this group of writers, but 
 also in sentence structure and word formation. 
 
 Latin Inscriptions found in Spain, 
 
 49 
 
 In the Christian inscriptions again, the influence of the Vul- 
 gate must, of course, be taken into account ; its restraining effect 
 is no doubt to be seen in the fixity of expression and literalness 
 that to some extent prevail. Often the thought seems a mere 
 copy instead of a flexible product of the writer's own mind. This 
 is sometimes sufficient to account for a strange use of the In- 
 finitive, a rare instance of the Subjunctive, or some other con- 
 struction at variance with the normal Latin idiom. 
 
 To one seeking or expecting abundant traces of the beginnings 
 of Spanish as distinct from the other Romance tongues, these 
 inscriptions are disappointing, for while they throw light upon 
 the development of the Romance languages in general, to the 
 development of Spanish in particular they have contributed 
 practically nothing outside of vocabulary. 
 
 '11 
 
 »1. c, 1, p. 126; cf. Bonnet, 1. c, p. 
 
 593. 
 
VITA. 
 
 Henry Martin was bom near Fredericksburg, Va., Nov. 26, 
 1878. His preliminary education was acquired in the schools 
 of Caroline County. His collegiate course was pursued at Eich- 
 mond College, Va., where he graduated with the degree of 
 Bachelor of Arts in 1899, and Master of Arts in 1901. During 
 four years he taught in secondary schools in the State. In 1903 
 he entered the Johns Hopkins University for graduate work, 
 selecting as his subjects Latin, Spanish and French. He also 
 completed the course in undergraduate Greek at the University 
 and attended for a year graduate lectures under Professors Gil- 
 dersleeve and Robinson. He was appointed Fellow in Latin in 
 1905 and in 1906 was Latin Instructor at Notre Dame Col- 
 lege of Maryland. Acknowledgments are due to the Professors 
 in the Romance Department of the University, to Prof. H. L. 
 Wilson, at whose suggestion this subject was undertaken, and to 
 Prof. K. F. Smith, whose many acts of kindness are gratefully 
 remembered. 
 

 /■ 
 
 / 
 
 s:? 
 
 t^M ■ 
 
 
 
 
 i*"' 
 
 I 
 
 COLUMBIA UNIVERSITY 
 
 0032190522 
 
 JAN 1^ n^ 
 
r 1