The spreadity of verbal prefixes

U radu se uvodi pojam proširenost kao jedne od mogućih brojčanih mjera koje mogu poslužiti za proučavanje značenja glagolskih predmetaka, posebno uporabnoga. Najprije se određuju pojmovi plodnost, čestotnost i proširenost i njihove se razlike oprimjeruju na maloj građi. Potom se donose brojčani podatci o glagolskim predmetcima koji pokazuju njihovu proširenost. Glavna je građa utemeljena na popisu pet tisuća hrvatskih glagola odabranih prema čestoti, što znači da su uključeni svi najčešeći u opsegu u kojemu je čestota važna. Taj je popis proširen predmetcima, čime se dobio glavni popis, gotovo dvostruko veća građa. U njoj se raščlanjuju brojčani podatci o proširenosti plodnih, slabo plodnih i neplodnih predmetaka poznatih iz literature. Pokazuje se i da prijedložni predmetci imaju znatno veći udio među glagolskim predmetcima nego (danas) neprijedložni predmetci, u čemu su slični izrazno jednakim najčešćim hrvatskim prijedlozima.This paper introduces the concept of spreadity of affixes as one of the numerical measures which could be used in the analysis of verb meanings, especially in their usage. It also suggests the need to distinguish between individual and common frequency when the number of affixes are analysed in a certain corpus. Individual frequency includes the number of tokens of the most frequent word formed with a certain affix, while common frequency includes tokens of all words formed with the same affix. The first part of the paper discusses the concepts of frequency, productivity and spreadity. Those three terms are applied to the four verb prefixes that appear in 500 and 4000 most frequent words as examples of the meaning difference between them. E.g. prefix emph{ne-} and prefix emph{pred-,} both semi-productive, differ in their individual frequency as well as their common frequency in the corpus of the 4000 most common Croatian words. Prefix emph{ne-} is a part of the eight times more frequent individual verb (1215 : 142) than prefix emph{pred-} (individual frequency), it is five times more frequent (1696 : 336) when all verbs with both the prefixes in the mentioned corpus are included (common frequency), but its spreadity is lower than the spreadity of emph{pred-} (4 vs. 6).\ The second part of the paper presents data on verbal prefixes in 9079 Croatian verbs. The choice of verbs is based on the most frequent app. 5000 verbs and verbs derived from them by various prefixes. The spreadity index is compared to the previous notions on productive, semi-productive and non-productive prefixes to see if there are any parallels. The results show that prefixes derived from prepositions that are used in contemporary language have a much higher spreadity than other prefixes found in the formation of Croatian verbs. Also, there are strong parallels between the frequency order of verbal prefixes and the prepositions with the same forms, although the most prototypical prepositions: emph{u} `in\u27 and emph{na} `at, on\u27 did not win the first place. The first 15 on the list include only three (today) non-prepositional prefixes (emph{raz-, pro-, pre-}), the rest are used as prepositions too (emph{za-, iz-, u-, po-, o-, s-, na-, pri-, od-, do-, pod-, uz-}). In the last 14 the majority do not have prepositional pairs (emph{ob-, su-, obez-, re-, de-, dez-, dis-, protu-, mbox{ne-}),} only four have (emph{nad-, pred-, mimo-, suprot-}). First place is occupied by emph{za-} (941), it is followed by emph{iz-} (707), emph{u-} (700), emph{po-} (675), emph{o-} (552), emph{s-} (539)