CC BY
11Philological Studies 17,1, (2019) 305-330
Izvirni znanstveni clanek
UDK: 811.111'373.237:811.163.6 doi: 10.17072/1857-6060-2019-17-305-330
MORPHOLEXICAL PROPERTIES OF ENGLISH AND SLOVENIAN MASCULINE/FEMININE NOUNS FOR PROFESSIONS
Gasper Ilc Andrej Stopar
Faculty of Arts University of Ljubljana Ljubljana, Slovenia
Key words: associative meaning, collocators, corpus frequency, feminatives, masculinatives, word-formation, semantic preference.
Summary: The article presents an analysis of masculine and feminine nouns denoting professions in English and Slovenian. The researched expressions are first discussed from the point of view of word-formation (derivation and compounding), then corpora are employed to examine the frequency of the forms in both languages. The corpus data are complemented with an analysis of collocators with the purpose of identifying the semantic preferences and associative meanings of gender-marked expressions for professions. The results reveal that some feminine nouns for professions are stylistically unmarked (especially in Slovenian), whereas others co-occur (in English and Slovenian) with words that reflect the societal attitudes to gender roles, appearance and character.
MORFO-LEKSIKALNE ZNACILNOSTI ANGLESKIH IN SLOVENSKIH MOSKIH/ZENSKIH SAMOSTALNIKOV ZA POKLICE
Gasper Ilc Andrej Stopar
Filozofska fakulteta Univerze v Ljubljani Ljubljana, Slovenija
Kljucne besede: asociativni pomen, feminativi, kolokatorji, maskulinativi, oblikotvorje, pogostnost v korpusu, semanticna preferenca.
Povzetek: Clanek predstavi kontrastivno analizo moskih in zenskih samostalnikov za poklice na jezikovnem paru anglescina-slovenscina. Obravnavani izrazi so najprej predstavljeni z besedotvornega vidika, prek izpeljave oziroma zlaganja. Prispevek nato s pomocjo korpusov preverja pogostnost opazovanih oblik v obeh jezikih. Zbrane korpusne podatke razcleni tudi s pregledom njihovih kolokatorjev, s cimer opredeli semanticne preference in pomenske asociacije obravnavanih feminativov in maskulinativov. Rezultati pokazejo, da so feminativi za poklice lahko nezaznamovani (sploh v slovenscini) ali pa se pojavljajo z izrazi, ki (v anglescini in v slovenscini) odsevajo druzbeni odnos do spolnih vlog, videza in osebnostnih lastnosti.
1 Introduction
It is well-known that several distinct levels of (un)markedness can be observed with nouns specifically denoting male or/and female referents. Starting with morphology, morphologically marked nouns for gender involve (un)markedness in terms of the (non-)presence of special gender-related morphology, be it inflectional or derivational. In English, for example, the noun actor is morphologically unmarked as it lacks any gender-related morphological markings, whereas the gender-related morpheme -ess in actress makes the female-referring noun marked. Exactly the same can be observed for their Slovenian counterparts igralec (masculine noun, male or female referents, morphologically unmarked) and igralka (feminine noun, female referents, morphologically marked) 1. At the level of semantics, it has traditionally been assumed that in a pair of masculine/feminine nouns (e.g., actor/actress), the masculine noun is unmarked as it can denote both female and male referents, whereas the feminine noun is marked, allowing only the denotation of female referents (cf. Huddleston in Pullum, 2002: 1680ff; Toporisic, 2000: 266; Vidovic-Muha, 2018). The current usage, however, shows that the unmarkedness of masculine nouns for both genders is disputable (Kranjc in Ozbot, 2013: 234). The reason for this development can lie in the fact that it is often the
1 This is a slight oversimplification. Both nouns are derived from the verbal base igral-, and then the derivational morpheme -ec is added for the masculine noun, and -ka for the feminine noun (cf. Vidovic-Muha, 2018: 410). The morpheme -ec carries no overt masculine marker, but this is the case of masculine nouns of the 1st declension, whereas the morpheme -ka carries the overt feminine marker a, typical of feminine nouns of the 1st declension (cf. Vidovic-Muha, 2018: 415).
case that the feminine noun has acquired some negative connotations through usage, so the users prefer the non-connotative masculine noun instead (e.g., poet/poetess, master/mistress). This observation leads us to the third type of markedness, stylistic, which is the focus of the present paper. In particular, the paper addresses the question of the stylistic (un)markedness of English and Slovenian gender-marked expressions for professions (henceforth: masculine/feminine nouns), analysing the data from different referential corpora: BNC, English web corpus 2013 (enTenTen13), and the Brown Family of English, and FidaPLUS v2 for Slovenian.
The corpus-based approach adopted herein examines gender-marked expressions for professions in their typical (co-)texts. The data on the cooccurrence of lexical units in corpora are used to analyse various aspects of their use. In line with Sinclair (1996), our choice of units of meaning depends on their linguistic co-text - his term 'extended units of meaning' thus takes into account the attractions between the word and its typical environments, which are based on collocation, colligation, semantic preference and semantic prosody.
Collocations are frequent co-occurrences of lexical units; Stubbs (2001) exemplifies such combinations with the phrase rancid butter. Gabrovsek (2005: 210-12) points out that collocations can be either semantically or lexically motivated - a semantically acceptable combination may still be unacceptable with regard to the frequency of its use (compare *glacial age and ice age). More recent definitions of collocations consider several properties of collocations to define them as "lexically and pragmatically constrained recurrent cooccurrences of at least two lexical items which are in a direct syntactic relation with each other" (Bartsch, 2004: 76). A colligation, on the other hand, refers to the frequent co-occurrences of words based on the syntactic patterns that include, for instance, negation, mood, person, number, tense, and quantifiers. For example, the verb deny requires a gerund, whereas the verb decline requires an infinitive: He denied visiting him. vs. He declined to visit him.
Corpora enable us to determine the discursive (or lexical/semantic) fields typical for a unit of meaning (Stubbs, 2007: 178; McIntyre, 2018: 87). The noun saucer, for example, is frequently found next to teapot, teacup, plate, and drop, but also next to flying (BNC via Sketch Engine, 2019). Saucer thus often appears in texts that describe tea drinking, setting the table, and similar; however, the same word is also frequent in texts related to unknown flying objects. In some cases, these word combinations can be classified as proper collocations (matching saucers), while in others they do not correspond to our initial definition of collocations. Such an example can
be seen in the clause to stack the dirty cups and saucers on the tray: here the nouns saucers and tray do not represent a collocation, but rather exhibit a tendency to appear in the same text. Stubbs (2001) presents a similar case by discussing the co-occurrence of large and various words denoting quantity (number, amount, scale). Gabrovsek (2005, 211) explains that such combinations have been classified by Crowther, Dignen and Lea (2002) as examples of categorial collocations - frequent co-occurrences within the discursive field. The latter are also closely related to the term semantic preference (Stubbs, 2001 and 2007), which is the tendency of the word to co-occur with words that belong to the same semantic set. Philip (2010) claims that semantic preference is shared by all speakers of a language.
Closely related to semantic preference is Hoey's (2005) associative meaning, which focuses on the wider context and on the process of learning a word that includes acquiring its referential meaning, syntactic environments, connotations, and associations. Speakers expect the lexical unit to behave in a certain way and associate it with a set of meanings that (may) entail its collocates (Hoey, 2005: 26). The difference between Hoey's associative meaning and Sinclair's semantic preference is in that the former also entails the secondary meanings of associations, evaluations and connotations. Philip (2010) describes this as a psychological aspect of understanding the lexical unit which is not necessarily shared by all speakers. For instance, the frequent use of adjectives describing appearance (beautiful) when discussing women, and the use of adjectives expressing importance (leading) when discussing men may reflect value judgments with potential societal ramifications (Caldas-Coulthard and Moon, 2010).
When the semantic aspects of lexical units are assigned a pragmatic function, semantic prosody comes into play (Philip, 2010). McIntyre (2018: 87-88) explains that, according to Sinclair, "semantic prosody describes the discourse function of the unit of meaning" and emphasizes that semantic prosodies "belong to units of meaning and not to individual words". He discusses the nouns job and career and shows that they both convey a similar semantic preference, yet corpus hits show that the typical collocates of career more often express a positive semantic prosody. In this respect semantic prosody refers to the positive and negative associations acquired by neutral words when they co-occur with other words (Sorli, 2012 and 2013; Gabrovsek, 2014; Jurko, 2015).
The sections that follow use the extended units of meanings - especially semantic preference and associative meaning - to analyse the corpus data for masculine/feminine nouns for professions.
2 Gender-marked nouns for professions: basic theoretical tenets
With a few exceptions (see below), English allows only so-called notional or semantic gender, meaning that the noun acquires its gender through reference with a real person (cf. Quirk et al., 1999: 314; Huddleston and Pullum, 2002: 484). For instance, the noun teacher is morphologically unmarked for gender, and displays semantic gender agreement with its referent (1a-b). When there is no referent (i.e., generic reference), the noun can refer to either male or female referents or both (1c).
(1) a) She was a very popular teacher because of her personal interest in them.
b) I'm not sure what my dad does. I think he is a teacher in the Air Force.
c) This is a first-level administrative post within an education department, usually for an experienced teacher. He or she works under an assistant education officer2.
In some cases, English has special forms to distinguish masculine/feminine nouns. These can be morphologically unmarked pairs (e.g., boy/girl, fox/vixen), or morphologically marked pairs3. The feminine nouns are derived either by the gender-marking morphemes -ess, -ette, -ine, -trix, or by compounding with the component elements she, woman, whereas the masculine nouns by the gender-marking morpheme -er, or the gender-marking component elements he and man (cf. Quirk et al., 1999: 1546f.; Huddleston in Pullum, 2002: 1680ff.; Bauer, 1996: 220f.).
For everyday usage, it is advisable (cf. Huddleston in Pullum, 2002: 1680) to avoid using morphologically marked feminine nouns unless their use is semantically or pragmatically motivated. For instance, the difference between a count and a countess or an abbot and an abbess is not just gender-related but also semantic, as they refer to differ societal and legal positions. In a similar fashion, the use of the feminine nouns actress and waitress is justified, as they also denote different theatre roles and job specifics respectively. Another reason for using a morphologically marked feminine noun is contextual: he married a rich heiress is not the same as he married a rich heir (Huddleston in Pullum, ibid.).
The main reason for avoiding some feminine nouns may lie in the fact that diachronically these nouns acquired negative connotative meanings and
2 All the examples listed in the paper are taken from the following corpora: BNC and English Web 2013 for English and FidaPLUS v2 for Slovenian.
3 There is also a third group of nouns, whereby the gender is fully grammaticalized and involves the personification of inanimate nouns (e.g., states, vehicles, emotions, etc.).
are consequently felt as pejorative in present-day usage (cf. Holmes, 2001: 126). Thus, derivates such as poetess, authoress and editorette, are of rare occurrences (see Table 1).
Table 1. poetess, authoress and editorette vs. poet, author and editor
query BNC enTenTen13 BNC enTenTen13
No. per M4 No. per M No. per M No. per M
poetess 25 0.22 1 762 0.08 poet 2 917 25.96 334 032 14.7
authoress 23 0.20 1 089 0.05 author 6 809 60.61 2 726 767 120.0
editorette 0 0 1 0 editor 5 440 48.42 1 316 870 57.9
In contrast to English, in Slovenian, gender is fully grammaticalized, meaning that the gender of a noun can either be the same of different from the gender of the referent. As a rule, the noun preserves its gender regardless of its referent (2a). There are, however, some exceptions. Some nouns such as staresina ('a chief) and vodja ('a leader')5 can be either feminine or masculine depending on the referent (2b,c) due to their morphological make-up (cf.: Uhlik, 2019).
(2) a) Ona je dober/*dobra pedagog. sheFEM is goodMAsc/*goodFEM educatorMAsc 'She is a good educator.'
b) Kot skrben vodja je priljubljen pri delavcih. as attentiveMASc leaderMASc is popularMASc at workers 'Being attentive, he is very popular among the workers.'
c) Vodja je prisluhnila nasim zeljam. leaderFEM is listenedFEM our wishes. 'The leader listened to our wishes.'
When analysing feminine derivatives, Vidovic-Muha (2018: 408) identifies the suffixes -a, -(ar)ka, -esa, -ica, -ina, -inja, -isa, -ja, -ka, -(ov)ka, -na and -ulja, with the following productive suffixes: -(ar)ka, -ica, and -inja. With regard to the suffixes used to derive nouns denoting male professions, the author (Vidovic-Muha, 2018: 410) discusses the suffix -ec. In her paper on designators for female holders of professions in Slovenian and Russian, Derganc (2017: 23f) argues that those in Slovenian are
4 Per million words.
5 Vodja can denote both male and female referents; however, its derivatives racunovodja ('an accountant') and poslovodja ('a manager') have feminine derivatives ending in -inja:
racunovodkinja and poslovodkinja, respectively.
stylistically unmarked. It is noteworthy that the corpus data show that in most cases the masculine nouns are much more numerous than their feminine counterparts (cf. Stumberger 2015). As already pointed out by Derganc (2017: 126), this is an expected observation, since, first, in the past the masculine nouns were used generically, and, second, the most prestigious positions and jobs were occupied by men. In addition to these two factors, special attention also has to be paid to the social context (cf. Vidovic-Muha, 2018: 413). For example, in Table 2, high values for the masculine noun zdravnik ('a doctor') can be attributed to the generic use, whereas the high values for the feminine nouns sobarica ('a chambermaid'), vzgojiteljica ('a kindergarten teacher'), and the masculine noun varilec ('a welder') result from the fact that these are the professions typically practised by females or males.
Table 2. Frequency of selected masculine/feminine nouns for professions in FidaPLUS v2
No. No.
zdravnik 34 775 zdravnica 4 811
ucitelj 17 842 uciteljica 8 782
sobar 111 sobarica 414
vzgojitelj 4 965 vzgojiteljica 6 535
varilec 418 varilka 9
3 Methodology
For the purposes of the analysis, we first extracted all English feminine nouns for professions from the British National Corpus, BNC, (100 million words, from 1960-1993, written and spoken discourse) in line with the morphological specifications presented in section 2. The final list after the extraction contains 39 feminine nouns for professions derived by the suffixes -ess, -trix and the component element woman. To determine the frequency of feminine nouns in English, the data from the BNC were compared with the data obtained from the English web corpus 2013, enTenTen13, (19 billion words, written language from the .uk web sites from 2013). The data from the Brown Family corpora (American and British English, divided into four subcorpora at 30-year intervals - 1931, 1961, 1991 and 2006) were used to examine the frequency and the usage tendencies from a diachronic perspective.
The basis for the analysis of Slovenian feminine nouns was the database by the Employment Service of Slovenia, available online at: https://www.ess.gov.si/ncips/opisi-poklicev. The database contains
masculine nouns only, so the next step was to check the existence of corresponding feminine nouns in the Dictionary of Standard Slovenian (SSKJ2 available at: fran.si), and in the FidaPLUS v2 corpus. The final list after the extraction contains 289 feminine nouns for professions derived by the suffixes -esa, -ica, -inja and -(ar)ka.
Stylistic (un)markedness was ascertained with the help of the collocation candidate tool provided by the Sketch Engine interface. For each investigated noun, collocation candidates were sought in the span from -5 to +5 words, and then the logDice measurement was examined. According to Rychly (2008), the logDice highest value is 14, with the value of 7 and above indicative of collocational association (cf. Gablasova et al., 2017: 168f).
The analysis of extended units of meaning is based on Word Sketch, a feature of Sketch Engine that provides a summary of a word's grammatical and collocational behaviour (Kilgarriff et al., 2014). The data collected in this way enabled the study of stylistic (un)markedness based on the concepts of semantic preference and associative meaning (Sinclair, 1996; Hoey, 2005; Philip, 2010).
4 Results
4.1 Corpus frequencies
In English, the highest proportion of feminine nouns for professions is derived by the suffix -ess (21), followed by the compounds with woman (15). Only one noun ending in -ette was identified (usherette), and two nouns ending in -atrix (aviatrix and prosecutrix). In Table 3, we present those feminine nouns that were found at least 200 times in the enTenTen13 corpus. It should be noted that due to the different sizes of the enTenTen13 and BNC corpora, only a value per million words is comparable. The data from the Brown Family subcorpora is comparable in terms of raw frequency, since all subcorpora contain the same number of words.
Table 3. Frequency of feminine nouns in English
feminine noun enTenTen13 BNC Brown Family
1931 1961 1991 2006
No. per M No. per M No. No. No. No.
actress 310 864 13.70 1 259 11.21 18 25 61 38
mistress 132 414 5.80 1 179 3.45 25 32 22 39
waitress 73 030 3.20 361 3.21 3 8 14 13
hostess 59 524 2.60 422 3.76 11 31 20 10
stewardess 13 665 0.47 97 0.86 4 3 9 7
headmistress 6 243 0.27 255 2.26 1 6 0 0
huntress 6 218 0.27 7 0.06 0 1 0 0
saleswoman 3 842 0.17 22 0.20 0 0 0 3
policewoman 3 834 0.17 228 2.03 0 3 0 2
laundress 1 818 0.08 13 0.12 0 0 1 0
washerwoman 958 0.04 22 0.20 1 0 0 0
anchorwoman 897 0.04 2 0.02 0 0 0 0
manageress 803 0.04 104 0.93 0 1 1 0
schoolmistress 789 0.03 26 0.24 2 0 0 0
sculptress 726 0.03 11 0.10 0 0 0 0
usherette 629 0.03 28 0.25 0 0 0 0
aviatrix 535 0.02 1 0.01 0 0 0 0
directress 529 0.02 1 0.01 0 0 0 0
fisherwoman 509 0.02 2 0.02 0 0 0 0
instructuress 347 0.02 6 0.05 1 0 0 0
ambassadress 297 0.01 1 0.01 0 0 0 0
conductress 202 0.01 12 0.11 0 0 0 0
The corpus data in Table 3 shows that the most commonly used feminine nouns for professions in English are the nouns actress and waitress, but the data from the Brown Family reveal that there was a decrease between 1991 and 2006 in the use of the noun actress. With all other feminine nouns ending in -ess, there is a decrease in usage. Special attention must be paid to the nouns mistress and huntress. The data in Table 3 suggest more frequent values in the enTenTen13 corpus then in others; however, we believe that the increase should not be attributed to the more frequent use of the two nouns when denoting female professionals. The noun mistress is polysemous as it may also denote a female lover. A better insight into the actual use of the noun mistress when denoting female professionals can thus be observed in the case of the compounds headmistress and schoolmistress: in both cases there is a sharp decrease in usage. The noun huntress, on the other hand, can also denote characters in computer games and films, so for this reason the occurrence of huntress is relatively high in the enTenTen13 corpus. There is a slight increase in the
use of some compounds with woman (e.g., anchorwoman, saleswoman); however, in some other cases, the usage is on the decrease (e.g.,
policewoman).
In Slovenian, the majority of feminine nouns denoting professions are derived by the suffix -(ar)ka (190), to be followed by the suffixes -ica and -inja (each around 50). Only two nouns with the suffix -esa were identified: klovnesa ('a clown') and stevardesa ('a stewardess'). In some cases, it may seem that the feminine nouns for the same profession may be derived by two different suffixes, for example, mesarica and mesarka. However, mesarica ('the butcher's wife'), according to the SSKJ2, does not denote a profession but a social position of a woman (cf. Vidovic-Muha, 2018: 408), whereas the noun mesarka ('a butcher') pertains to the profession, and has as such also been standardised in the National Qualifications Framework (SOK). The SSKJ2 defines the noun kovacica in a similar fashion as mesarica ('the blacksmith's wife'), but the SOK uses the same word for a female profession (i.e., 'a female blacksmith'). The masculine noun geograf ('a geographer') has two feminine counterparts geografka and geografinja, with only the latter being standardised in the SOK. In addition to female nouns, our analysis has identified the morpheme -ar, which is used to derive masculine nouns from feminine nouns (e.g., babicar 'a male midwife').
Table 4 shows those feminine nouns that have at least 200 occurrences in the FidaPLUS v2. The data are provided separately for the 1979-1990 and 200-2006 subcorpora. Due to different sizes of the subcorpora only the value per million words is comparable6.
Table 4. Frequency of feminine nouns in Slovenian
feminine noun subcorpus 1979-1990 subcorpus 2000-2006 total feminine noun subcorpus 1979-1990 subcorpus 2000-2006 total
No. Per M No. per M No. No. per M No. per M No.
igralka 9,091 45.00 37,246 72.00 46,337 sobarica 248 1.23 774 1.49 1,022
direktorica 6,724 33.00 25,340 49.00 32,064 gasilka 279 1.00 724 1.00 1,003
babica 4,916 24.00 15,374 30.00 20,290 izvajalka 203 1.00 796 2.00 999
uciteljica 4,628 23.00 11,924 22.90 16,552 fotograimja 152 1.00 817 2.00 969
delavka 3,931 20.00 10,066 19.00 13,997 ekonomistka 264 1.00 695 1.00 959
novinarka 3,242 16.00 9,458 18.00 12,700 varuhinja 203 1.01 740 1.42 943
sodnica 2,488 12.40 9,385 18.02 11,873 kozmeticarka 149 1.00 775 1.00 924
6 Table 4 also lists nouns babica, detektivka, kuharica, lovka and spremljevalka, even though these are polysemous items: babica ('a midwife' or 'a grandmother'), detektivka ('a female detective' or 'a detective book/film'), kuharica ('a female cook', or 'a cookery book'), lovka ('a female hunter' or 'a tentacle'), and spremljevalka ('a companion' or 'a runner-up in a beauty competition').
3,644 18.10 8,183 15.71 25 0.12 874 1.68
sekretarka 11,827 stilistka 899
3,448 17.20 7,424 14.30 132 1.00 742 1.00
tajnica 10,872 menedzerka 874
2,647 13.17 6,692 12.85 170 1.00 681 1.00
ravnateljica 9,339 negovalka 851
2,200 10.94 5,892 11.30 167 0.83 674 1.29
svetovalka 8,092 predavateljica 841
3,077 45.31 4,726 9.07 247 1.23 572 1.10
poslanka 7,803 zastopnica 819
1,818 9.04 5,922 11.37 258 1.28 554 1.06
zdravnica 7,740 opazovalka 812
1,392 7.00 5,986 11.00 178 0.89 630 1.21
manekenka 7,378 vodnica 808
plesalka 1,412 7.02 5,929 11.40 7,341 referentka 217 1.08 589 1.13 806
1,735 8.63 5,562 10.68 211 1.00 583 1.00
prodajalka 7,297 lektorica 794
2,013 10.01 4,488 8.62 219 1.00 569 1.00
vzgojiteljica 6,501 napovedovalka 788
1,215 6.04 4,789 9.19 230 1.14 530 1.02
voznica 6,004 uradnica 760
1,204 5.99 3,956 7.59 207 1.03 517 0.99
pomocnica 5,160 racunovodkinja 724
1,417 7.00 3,383 6.00 243 1.00 456 1.00
natakarica 4,800 etnologinja 699
1,038 5.16 3,663 7.03 121 0.60 578 1.11
spremljevalka 4,701 vojakinja 699
804 4.00 3,613 6.94 117 1.00 536 1.00
reziserka 4,417 dirigenka 653
936 4.66 3,211 6.16 117 0.58 530 1.02
sportnica 4,147 terapevtka 647
999 5.00 2,866 6.00 188 0.94 440 0.84
kuharica 3,865 sociologinja 628
240 1.19 3,430 6.58 148 1.00 476 1.00
tehnica 3,670 fzioterapevtka 624
677 3.37 2,334 4.48 186 1.00 428 1.00
odvetnica 3,011 bolnicarka 614
526 2.62 2,454 4.71 177 1.00 435 1.00
oblikovalka 2,980 inzernrka 612
714 3.55 1,928 3.70 109 0.54 484 0.93
sivilja 2,642 pediatrinja 593
590 2.93 1,872 3.59 152 0.76 391 0.75
psihologinja 2,462 upravnica 543
529 3.00 1,875 4.00 87 0.43 438 0.84
izvedenka 2,404 scenaristka 525
661 3.29 1,595 3.06 103 1.00 404 1.00
prometnica 2,256 maserka 507
440 2.19 1,790 3.44 114 1.00 384 1.00
strokovnjakinja 2,230 biologinja 498
502 2.50 1,680 3.23 133 0.66 361 0.69
prevajalka 2,182 posrednica 494
522 3.00 1,616 3.00 155 0.77 334 0.64
inspektorica 2,138 receptorka 489
546 2.72 1,461 2.80 113 0.56 363 0.70
pedagoginja 2,007 oskrbnica 476
326 2.00 1,612 3.00 134 0.67 319 0.61
asistentka 1,938 poslovodkinja 453
472 2.00 1,461 3.00 8 0.04 434 0.83
cistilka 1,933 vizazistka 442
359 2.00 1,396 3.00 154 0.77 276 0.53
mojstrica 1,755 strezmca 430
326 2.00 1,393 3.00 140 0.70 286 0.55
glasbenica 1,719 zobozdravnica 426
361 1.80 1,338 2.57 92 0.01 309 1.00
trenerka 1,699 kostumografnja 401
287 1.43 1,384 2.66 83 0.01 304 1.00
stevardesa 1,671 cvetlicarka 387
640 3.18 1,015 1.95 96 0.01 287 1.00
veleposlanica 1,655 arheologinja 383
424 2.00 1,211 2.00 133 0.66 231 0.44
lovka 1,635 tocajka 364
345 2.00 1,272 2.00 91 0.01 266 1.00
kustosinja 1,617 izdelovalka 357
287 1.00 1,327 3.00 61 0.01 293 1.00
frizerka 1,614 antropologinja 354
336 1.67 1,253 2.41 43 0.01 310 1.00
policistka 1,589 klovnesa 353
306 1.52 1,212 2.33 78 0.39 269 0.52
organizatorka 1,518 veterinarka 347
290 1.00 1,225 2.00 112 1.00 233 0.10
arhitektka 1,515 administratorka 345
392 1.95 1,037 1.99 62 0.01 270 1.00
pravmca 1,429 notarka 332
274 1.00 1,125 2.00 61 0.01 259 0.01
koreografinja 1,399 animatorka 320
347 1.73 1,028 1.97 53 0.26 242 0.46
zgodovinarka 1,375 revizorka 295
330 1.64 1,027 1.97 103 0.51 181 0.35
solistka 1,357 telefomstka 284
305 2.00 1,042 2.00 42 0.01 211 0.01
blagajnicarka 1,347 farmacevtka 253
204 1.00 1,049 2.00 108 1.00 125 0.01
detektivka 1,253 mlinarica 233
321 2.00 923 2.00 6 0.03 223 0.43
knjiznicarka 1,244 oskrbovalka 229
283 1.00 886 2.00 78 0.39 147 0.28
dramaturginja 1,169 zaloznica 225
275 1.37 816 1.57 31 0.01 185 0.01
upraviteljica 1,091 maticarka 216
328 2.00 758 1.00 72 0.01 142 0.01
kiparka 1,086 hisnica 214
Comparing the data from both suborpora, we can observe that the number of feminine nouns for professionals is either on the increase (e.g., igralka ('an actress'), novinarka ('a journalist'), reziserka ('a film director')) or it maintains its value over time (e.g., delavka ('a worker'), zgodovinarka ('a historian'), solistka ('a soloist')). There are only few examples where a decrease has been observed (e.g., sekretarka ('a secretary'), telefonistka ('a telephonist')). The reason for such declines could be that these professions are slowly disappearing.
4.2 Collocators and semantic preferences
Corpus data reveal the attractions between the word and its typical environments. The masculine/feminine pairs of nouns below present the most frequent combinations, with special reference to semantic fields and semantic preferences.
The English data are as follows.
I Actress/actor (1,259/3,600 hits in BNC): Using the BNC corpus and the Word Sketch function in Sketch Engine, it can be observed that some of the most frequent premodifiers of the feminine form are best, actor, bitpart, out-of-work, sought-after and singing. The word also co-occurs with nouns such as actor, singer, and dancer, and is often accompanied by names of well-known individuals (Hepburn). Among the verbs appearing in
316
its proximity are also star, marry, support, win, lead, watch and become. The masculine form actor, on the other hand, is frequently premodified by words such as character, best, Hollywood, talented, comedy and unemployed; furthermore, it can be followed by, for instance, actress, singer, musician, and award; and it also co-occurs with the names of individuals (Branagh). The verbs that co-occur with the masculine form include vote, lead, dress, become, name, concern, play and portray. A comparison of the most frequent combinations in EnTenTen13 (via Sketch Engine, with a span of +/-5 words) shows that the feminine form is noticeably more frequently combined with the modifiers porn and supporting.
considering the semantic preferences, both forms combine with expressions describing related professions; place names related to the profession; names of individuals; expressions describing the type/quality of work; activities related to the profession; and expressions related to awards. Some specific connotations of the feminine form are brought to the fore by the collocators such as aspiring, supporting, porn, affair and steamy; some of these may also appear with the masculine form, but less frequently.
II Policewoman/policeman (228/3,193 hits in BNC): The nouns policewoman/policeman have been largely replaced by the gender-neutral police officer; nevertheless, the data from BNc reveal some interesting differences between them. The feminine form appears after the premodifers like zoggian, plain-clothed, unlucky, probationer, off-duty and daredevil. The word is often followed by nouns such as baillet, teleanou, field-worker and firewoman; it also appears in the proximity of words like radio, rape, gun, fancy, staff and retreat. The most frequent collocators (based on the logDice value) of the feminine form are also unluckiest, smartly and brave, which do not appear next to the masculine form. The masculine form policeman often co-occurs with off-duty, neighbourhood, plain-clothes, sleeping, secret and ordinary; it is used with verbs such as injure, assault, kill, patrol, arrest and kick; and it co-occurs with nouns that frequently include civilian, soldier, woman, fireman, detective and policewoman.
Regarding semantic preferences, both forms appear with expressions that refer to police activities or describe individuals connected to police work. The feminine noun is more frequently found next to personal names, in sci-fi contexts, and with adjectives that point to (positive) personal traits (smartly, brave). The observed word rape refers to an attempted rape of a female police officer.
III Waitress/waiter (361/932 hits in BNC): The feminine form waitress is often premodified by cocktail, 17-year-old, assistant, Russian, head, young and good. It co-occurs with nouns such as job, service, waiter, chambermaid and cleaner, and verbs like call, ask, bring, be, serve and arrive. Among the frequent collocators are also the words Hooters, topless and flirt (logDice: 6.6, 5.8 and 5.3 respectively). The masculine noun appears with the premodifiers tide, obsequious, white-jacketed, head, dumb and wine. Waiter also co-occurs with waitress, attendant, chef, barman, maid and menu. It can also be found next to beckon, hover, summon, bow, serve, bring, pour and similar verbs.
With respect to semantic preferences, the pair waitress/waiter attracts collocators referring to related professions, typical activities, workplace objects, and expressions related to the activities of customers. Both also frequently co-occur with expressions related to appearance and character in which waitresses are more frequently positively perceived (attentive), and waiters negatively (dumb). The feminine noun also co-occurs with expressions that sexualize the profession (topless).
Moving to the Slovenian data, the feminine nouns can be divided into two main groups: Group A consists of those feminine nouns that do not differ from the masculine ones and are stylistically unmarked, whereas Group B includes those whose meaning is specialized or at least partly stylistically marked. The authors see these groups as extreme ends of a continuum, which is supported by examples that exhibit the features of both groups.
Representing Group A are the masculine/feminine pairs: zdravnica/zdravnik ('a female/male doctor'), sodnica/sodnik ('a female/male judge'), and vzgojiteljica/vzgojitelj ('a female/male kindergarden teacher' ). Group B includes the pairs administratorka/administrator ('a female/male administrator'), tajnica/tajnik ('a female/male secretary'), kuharica/kuhar ('a female/male cook'), maserka/maser ('a female/male masseuse/masseur'), and cistilka/cistilec ('a female/male cleaner'). The selection of examples aims to present both prototypical and borderline examples in each group.
Group A (semantically equivalent, stylistically unmarked forms):
A-I Zdravnica/zdravnik ('female/male doctor'; 7,857/127,248 hits in FidaPLUS v2):7 The most frequent collocators of the feminine noun (logDice above 7, span +/-5 words) are dezurna ('on-call'), specialistka
7 The number of hits stated here differs slightly from the one in Table 4; this is due to the sub-corpora not containing all the data from FidaPLUS v2.
('specialist'), pregledala ('examined'), medicine and Quinn - the results are slightly skewed here because of the series Dr. Quinn, Medicine Woman appearing in the numerous TV listings included in the corpus. The function Word Sketch, on the other hand, lists collocators that include the adjectives dezurna ('on-call'), oddelcna ('departmental'), cenjena ('highly valued'), zavodska ('institutional'); nouns such as znanstvenica ('a scientist'), pravnica ('a jurist'), upraviteljica ('a manager'), pedagoginja ('a pedagogue'); and verbs like oskrbeti ('treat'), predpisati ('prescribe'), prihiteti ('hurry'), odrediti ('order'), and pregledati ('examine'). The masculine form co-occurs with the adjectives dezurni ('on-call'), splosni ('general'), izbrani ('chosen'), hribovski ('mountain'), druzinski ('family'); nouns like zobozdravnik ('a dentist'), bolnik ('a patient'), sestra ('a nurse'), zdravnica ('a doctor'); and verbs that include predpisati ('prescribe'), svetovati ('advise'), priporocati ('recommend'), pregledati ('examine'), and oskrbeti ('treat'). Some of the most frequent co-occurring expressions (logDice above 7, span +/-5 words) are pojdite ('go'), predpise ('prescribe'), dezurni ('on-call'), zaupajo ('trust'), splosni ('general'), bolniki ('patients'), sestre ('nurses').
The semantic preference for both forms is as follows: adjectives for the type of work done by doctors; verbs describing decisions related to medical treatment; and related professions.
A-II Sodnica/sodnik ('female/male judge'; 11,946/115,838 hits in FidaPLUS v2): The feminine noun frequently (logDice above 7, span +/-5 words) appears in TV listings, which is evidenced by the frequent collocator raztresena ('absent-minded'), a word used in the Slovenian title of the series Ally McBeal. Similarly troublesome for the analysis are those hits referring to the series Judge Judy and Judging Amy, since the Slovenian translations of their titles include the word sodnica ('female judge'). More insight can be gained by examining the Word Sketch data. The frequent modifiers of the feminine noun include preiskovalna ('investigative'), okrajna ('county'), razpravljajoca ('hearing'), and dezurna ('on-call'). Female judges preside (predsedovati), order (odrediti), judge (razsoditi), grant (ugoditi), and interrogate (zaslisati). The feminine noun also appears next to nouns such as odlocitev ('decision'), vprasanje ('question'), nepristranskost ('impartiality'), toznica ('claimant'), odvetnica ('lawyer'), and pravnica ('jurist'). Some of the frequent collocators also include words related to sports: strelska ('shooting'), trenerka ('coach'), turnir ('tournament'). The findings are very similar when it comes to the masculine form: ustavni ('constitutional'), predsedovati ('preside'),
izvolitev ('election'), porota ('jury'), nogometni ('football'), tekma ('match'), etc.
Both masculine and feminine forms appear in contexts related to law and sports. The semantic preferences are as follows: the position within the profession; activities related to decision-making; and related professions. A comparison of adjectives premodifying the two forms reveals a significant difference: the adjectives denoting important roles within the profession are more frequently found next to the masculine form: ustavni ('[masculine] constitutional') - 11,042 hits / ustavna ('[feminine] constitutional') - 442 hits; vrhovni ('[masculine] supreme') - 2781 hits / vrhovna ('[female] supreme') - 253 hits; okrajni ('[male] county') - 491 hits / okrajna ('[feminine] county') - 654 hits. The finding is relevant from the point of view of associative meaning and extra-linguistic circumstances, which may also explain the attraction of female judges (and doctors) for TV produces/audiences.
A-III Vzgojiteljica/vzgojitelj ('female/male kindergarten teacher'; 6,535/4,965 hits in FidaPLUS v2): The word vzgojiteljica often co-occurs (logDice above 7, span +/-5 words) with the following collocators: vrtec ('kindergarten'), uciteljice ('teachers'), pomocnice ('assistants'), predsolski ('pre-school'), mentorica ('mentor'), otroci ('children'), opravili ('done'), poskrbele ('made sure'), etc. Word Sketch lists the typical adjectival promodifiers such as predsolska ('pre-school'), prizadevna ('dedicated'), diplomirana ('graduate'), usposobljena ('qualified'), zaposlena ('employed'), upokojena ('retired'); some of the verbal collocators are poucevati ('teach'), uresniciti ('realize'), razloziti ('explain'), spodbujati ('encourage'), skrbeti ('care'), zaposliti ('employ'), ustvarjati ('create'), nacrtovati ('plan'); while the nominal collocators include uciteljica ('teacher'), vzgojitelj ('kindergarten teacher'), pomocnica ('assistant'), varuhinja ('guardian'), ravnateljica ('principal'), stars ('parent') and others.
The masculine noun frequently appears with ucitelji ('teachers'), predsolskih ('pre-school'), otrok ('children'), vzgojiteljice ('kindergarten teachers'), starsi ('parents'), vrtci ('kindergartens'), pedagogi ('pedagogues'), etc. Other collocators include dijaski dom ('student dormitory'), skrbnik ('guardian'), gojenci ('boarders'). The hits thus reveal a strong link between the masculine noun and secondary-school dormitories and similar institutions, which is confirmed in Word Sketch that lists adjectives like usposobljeni ('qualified'), pozorni ('attentive'), dezurni ('on-call'), strog ('strict'), zasebni ('private'); verbs such as pretepsti ('beat'), vzgajati ('raise'), spodbujati ('encourage'), poucevati ('teach'), skrbeti ('care'), sprasevati ('ask'); and nouns like vzgojiteljica ('kindergarten
teacher'), ucitelj ('teacher'), katehet ('catechist'), stars ('parent'), gojenec ('boarder'), skrbnik ('guardian'), solnik ('educator'), etc.
The described corpus data do not reveal any substantial differences between the masculine and feminine nouns: both are similarly frequent and appear with similar collocators. The semantic preferences can be described as: relating to work environment; people in related professions; and activities typical for the profession. Both forms also co-occur with similar expressions that describe an individual's character and their professional qualifications. Nevertheless, the data suggests that some elements of gender stereotyping may be present: for instance, female kindergarten teachers are dedicated, while male ones are strict. A similar finding can be associated with the working environment of student dormitories, where the masculine form is more common.
Group B (narrower semantic meaning, partial stylistic marking):
B-I Administrator/administrator ('female/male administrator'; 345/2,876 hits in FidaPLUS v2): The SSKJ2 defines the feminine form as "a lower office worker" and as "a woman who administers webpages". The masculine form is defined as "one who manages, a manager"; the same definition includes the use of the word in the field of information technology ("one who manages a computer system or an online forum") and, as the last option, the meaning of "a lower office worker". The staterun Employment Service of Slovenia only lists office-related positions in connection with this job title.
The corpus data (via Word Sketch) reveal that the feminine noun appears next to adjectives such as notarjeva ('notary's'), nezaposlena ('unemployed'), policistova ('policeman's'), radodarna ('generous'), ljubezniva ('kind'), vestna ('conscientious'), and postavna ('shapely'); verbs like pridirjati ('come running'), ozigosati ('stamp'), and prepisovati ('copy'); and nouns that include pisanje ('writing'), streznica ('hospital attendant'), racunovodkinja ('accountant') and menedzerka ('manager'). The masculine form is found next to adjectives such as apostolski ('apostolic'), sistemski ('systemic'), mrezni ('network'), dekanijski ('deanery's'), kolonialni ('colonial') and sposobni ('competent'); verbs like pregledovati ('check'), dodeljevati ('assign'), upravljati ('manage') and vzdrzevati ('maintain'); and nouns such as forum ('forum'), spica ('peak'), streznik ('computer server'), povezava ('connection'), baza ('database'), moderator ('moderator'), teoretik ('theoretician') and programmer ('programmer'). The typical collocators (logDice above 7, span +/-5 words) show that the most frequent ones involve expressions related to IT, for instance names, nicknames and other expressions common for online
content: moderator ('moderator'), pridruzen ('joined'), timi666 (a nickname), sistemski ('system'), sporocilo ('message'), etc.
The semantic preference of the feminine form can be identified in its cooccurrence with professions denoting superior positions; (gentle) character traits; appearance; office-related activities and locations. The masculine form collocators confirm its semantic preference for management positions, expressions describing high professional qualifications, and administration in the field of IT.
B-II Tajnica/tajnik ('female/male secretary'; 11,016/12,931 in FidaPLUS v2): The SSKJ2 defines the feminine form as "one who conducts administrative, organizational and technical business". The definition of the masculine form is identical, but it also includes the additional meaning of "a high official of some organization, body". The corpus-based analysis reveals, however, that the feminine form often combines with glavna ('main/head') in the sense of "high official". Despite this, the classification of occupations provided by the Employment Service of Slovenia lists the feminine form only in connection with its narrower meaning, and even states that the position is more commonly filled by women.
The frequent collocators for tajnica (Word Sketch) are telefonska/avtomatska ('phone/automatic') - referring to an answering machine - direktorjeva ('director's'), ministrove ('minister's'), zupanova ('mayor's'), Clintonova ('Clinton's') and similar. The word combines with verbs such as prevezati ('transfer calls'), se oglasati ('answer'), povedati ('tell'), vstopiti ('enter'), obvescati ('inform'), poklicati ('call'), and z njimi varati ('cheat with them'). It stands in proximity to nouns like izpopolnjevanje ('training'), klub ('club'), kongres ('congress'), stol ('chair'), pisarna ('office'), fikus ('ficus'), racunovodkinja ('accountant'), blagajnicarka ('treasurer'), receptorka ('receptionist'), snazilka ('cleaner'), ljubica ('lover'), razmerje ('affair'), etc. Corpus hits include sentences describing a diplomat who obtained an office, a ficus plant and a secretary, or bankers who are afraid to lose their armchairs, substantial salaries, secretaries and ficus plants.
The most frequent collocators of the masculine form (logDice above 7, span +/-5 words) are prominent functions in organizations or names of individuals occupying these positions: glavni ('head/main'), sindikata ('union's'), sinode ('synod's') and demokratov ('Democrats''). According to Word Sketch, the form tajnik is often close to adjectives such as glavni ('main'), vatikanski ('Vatican'), generalni ('general'), and papeski ('papal'). The typical verbs in the proximity of the masculine form are sluzbovati ('serve'), povedati ('tell'), pojasniti ('explain'), zatrditi
('assure'), and nagovoriti ('address'). The masculine form also appears near nouns like blagajnik ('treasurer'), podtajnik ('deputy secretary'), podzupan ('deputy mayor'), racunovodja ('accountant'), knjiznicar ('librarian'), as well as near names of well-known personalities.
Regarding semantic preference, we can conclude that the feminine form is frequently linked to the business space led by an influential male (director) and surrounded by objects typical for such a space (office plants). In addition, the activities that tajnica is involved in are related to the flow of information and involve other professions below the managerial level. The examples clearly reflect the associative meaning of the feminine noun (a secretary is objectified by being equated with furniture and office plants). Some additional connotations can be identified in contexts related to socially less desirable interactions (affairs). On the other hand, the masculine form is mostly connected with the position of a high official. It co-occurs with important institutions, activities that involve addressing larger groups of people, and related professions.
B-III Kuharica/kuhar ('female/male cook'; 3,907/14,892 hits in FidaPLUS v2): The form kuharica is polysemous - besides the frequent meaning of "a woman who is (professionally) occupied with cooking", the meaning of "cookbook" is also common (see the SSKJ2). The masculine form is defined as "one who is (professionally) occupied with cooking".
The list of collocators for kuharica (logDice above 7, span +/-5 words) includes words such as Kalinsek (a well-known cook-book author), zaposlimo ('[we] employ'), Felicita (a well-known cook-book author), kuharji (' cooks'), natakarica (' waitress' ), izvrstna (' excellent'), Vendelina (a well-known cook-book author), cistilka ('a cleaning lady'), odlicna ('excellent'), recepti ('recipes'), and sobarica ('a chambermaid'). The function Word Sketch lists adjectives like farovska ('parochial'), ravnateljeva ('principal's'), izvrstna ('excellent'), zlehtna ('wicked'), izucena ('educated'), zupnikova ('priest's'), vedozeljna ('eager to learn'), and pozabljiva ('forgetful'). The feminine noun is often coordinated with the nouns cistilka ('a cleaning lady'), kuhar ('a cook'), natakarica ('a waitress'), sobarica ('a chambermaid'), and hisnik ('a caretaker'). Some examples of the most frequent words that co-occur with the feminine noun are menza ('canteen'), gostilna ('inn'), kuhinja ('kitchen'), vrtec ('kindergarten'), dom ('home'), recept ('recipe'), praksa ('practice'), izkusnja ('an experience'), and druzina ('family'). Among the verbal collocators the following can be observed: peci ('bake'), kuhati ('cook'), zaposliti ('employ'), obleci ('wear'), and upokojiti ('retire').
The masculine noun co-occurs with vaja ('an exercise'), natakar ('a waiter'), gospa ('Ms.'), hisa ('house'), nedolocen ('permanent'), najboljsi ('the best'), dragocenosti ('valuables'), and porocila ('reports'). Word Sketch lists the following premodifying collocators: pojoc ('singing'), ljubiteljski ('amateur'), ladijski ('ship's'), izvrsten ('excellent'), dvoren ('court'), izucen ('trained'), pozabljiv ('forgetful'), and vesc ('skilful'). Kuhar often stands close to the nouns like natakar ('a waiter' ), kuharica ('a cook'), slascicar ('a confectioner'), jedec ('an eater'), gurman ('a gourmet'), menza ('a canteen'), restavracija ('a restaurant'), kuhinja ('a kitchen'), hotel ('a hotel'), gostilna ('an inn'), pomocnik ('a helper'), kapa ('a hat'), sekcija ('a section'), ekipa ('a team'), tekmovanje ('a competition'), and tecaj ('a course'). The verbs frequently found next to kuhar are servirati ('serve'), izmisljati ('make up'), speci ('bake'), skuhati ('cook'), odrezati ('cut off'), izobrazevati ('educate'), postreci ('serve'), razjariti ('enrage'), najeti ('hire'), pripravljati ('prepare'), pokvariti ('spoil'), zaposliti ('employ'), andpohvaliti ('praise').
The semantic preferences for both nouns include expressions for related professions; character traits; professional qualifications; types of work; work environments; and activities related to the occupation. The analysis also reveals several socially constructed differences. The feminine noun is present in some specific work environments reserved for female cooks (parish house), and it appears with professions that typically require lower qualifications (cleaner). The masculine form, on the other hand, can refer to a ship's cook; it is more frequently described positively (excellent, trained); and is often associated with competitions (section, team, competition).
B-IV Maserka/maser (masseuse/masseur; 512/1,216 hits in FidaPLUS v2): The SSKJ2 uses the same definition for both forms: "a woman/one who professionally massages". Word Sketch lists some frequent premodifiers such as tajska ('Thai'), spretna ('skilful'), izurjena ('trained'), prikupna ('cute'), diplomirana ('graduate'); verbs zmasirati ('massage'), obdelovati ('do'); and nouns maser ('a masseur'), aromaterapevtka ('an aromatherapist'), fizioterapevtka ('a physiotherapist'), psihologinja ('a psychologist'), serviser ('a repairman'), sestra ('a nurse'), roka ('a hand'), etc. The corpus also includes the names of well-known sports masseuses. Some less frequent collocators (logDice under 7, span +/-5 words) of the feminine noun include eroticna ('erotic'), nezna ('gentle'), carobna ('magic'), oblecena ('dressed'), and ljubezen ('love'). The data for the masculine form are very different. According to Word Sketch, a masseur is tajski ('Thai'), izsolan ('educated'), izurjen ('skilled'), izkusen ('experienced), usposobljen ('trained'), ukrajinski ('Ukrainian'),
profesionalen ('professional'), and priznan ('renowned'), but not prikupen ('cute'). The most frequent verbs next to maser include pregnete ('knead'), gladi ('stroke'), masira ('massage'); it co-occurs with the nouns like pediker ('a pedicurist'), fizioterapevt ('a physiotherapist'), mehanik ('a mechanic'), kozmeticarka ('a beautician'), serviser ('a repairman'), psiholog ('a psychologist'), zdravnik ('a doctor'), and trener ('a coach'). The most frequent hits do not include the adjective eroticni ('erotic'). In addition, the masculine noun is more prevalent in the fields of sport and medicine.
The semantic preferences shared by both the masculine and feminine nouns can be identified in expressions related to professional qualifications; typical activities; and similar professions. The differences can be detected in the words describing personal traits and appearance (cute), as well as in corpus hits involving erotic massages - such collocators co-occur with the feminine form and testify to the often perceived sexualization of the profession.
B-V Cistilka/cistilec ('female/male cleaner'; 1,950/1,767 hits in FidaPLUS v2): Word Sketch lists the feminine form with adjectives such as optimisticna ('optimistic'), pridna ('hard-working'), bosanska ('Bosnian'), priljubljena ('popular'), mlada ('young'); verbs like pospravljati ('tidy up'), zboleti ('fall ill'), prezivljati ('support'), pobirati ('pick up'), cistiti ('clean'), zaposliti ('employ'); nouns as cistilec ('a cleaner'), kuharica ('a cook'), vrtnarka ('a gardener'), zaposlitev (employment), placa ('a salary'), and ciscenje (cleaning). The most frequent collocators (logDice above 7, span +/-5 words) include words like Fata (a female Bosnian name), sobarica ('a chambermaid'), and hisnik ('a caretaker'). The masculine noun most frequently stands next to words that suggest that it refers to cleaning appliances: visokotlacni ('high-pressure'), kemicni ('chemical'), parni ('steam'), vakuumski ('vacuum'), etc. Only a few collocators refer to people, e.g. a sentence in the corpus states that a shop is "meant for professional and amateur [male] cleaners, as well as housewives". The co-occurring verbs are izobrazevati ('educate'), cistiti ('clean'), zaposliti ('employ'), while the nouns include smetar ('a garbage collector'), vzdrzevalec ('a maintenance worker'), cistilnica ('a laundry shop'), kuhar ('a cook'), natakar ('a waiter'), traktor ('a tractor'), andprah (dust).
Considering the semantic preference, we can conclude that it encompasses words referring to character traits; typical activities related to cleaning; and similar professions that require lower qualifications. The associative meaning of the feminine noun can be observed in the hits that involve the collocators bosanska (Bosnian) and Fata (a stereotypical female
Bosnian name). The masculine form mostly refers to appliances, signalling that the profession of a cleaner is mostly perceived as a female one.
5 Discussion
The first part of the paper is focused on the derivation of feminine and masculine nouns for professions in English and Slovenian. The analysis shows that the dominant form of feminine nouns in both languages involves derivation via affixation, while English also features compounding. Masculinatives are rare in both languages; they are mostly formed in relation to professions which have been traditionally considered part of the female sphere. Besides the masculine suffix -ec, which was previously discussed by Vidovic-Muha (2018:410), the analysis also identifies the Slovenian masculine suffix -ar. The suffix is not productive but can nevertheless be observed in the word babicar ('male midwife'). The most notable difference between the two languages is that contemporary English seems to be abandoning the use of gender-marked forms (except for the rare feminatives that remain), whereas in Slovenian, as evidenced by corpus data, the use of feminatives is on the rise. The reason for this asymmetry is rooted in the fact that English has an almost non-existent grammatical gender system, and in most cases uses the same words to describe a man or woman working in a certain profession. In contrast, Slovenian grammatical gender is fully grammaticalized and may or may not overlap with biological sex. Slovenian nouns denoting professions have clear referents with regard to gender, and their contemporary use - evidenced in corpora - shows a clear tendency towards the agreement between grammatical and biological genders. Thus, in the case of female referents the use of feminine nouns is now preferred, while the use of masculine generic forms is on the decline.
Our corpus analysis shows that the differences between masculine and feminine nouns for professions are best revealed by examining their collocators, semantic preferences, and semantic associations (Sinclair 1996 and Philip 2010).
The collocators of English masculine and feminine nouns for the examined professions (actress/actor, policewoman/policeman in waitress/waiter) reveals that they share some semantic environments: related professions; places and surroundings associated with the profession; personal names; words describing the quality of their work; and the typical actions associated with the profession. Some specific associative meanings are also revealed if we examine the collocators describing human traits and appearance. The feminatives often co-occur with expressions for stereotypically female character traits and personality (including sexual
connotations). This is in line with Caldas-Coulthard and Moon (2010), who explain such asymmetries with stereotypes of gender, sexualization of women, as well as with age and behaviour.
Slovenian masculine/feminine pairs for professions reveal similar findings. Regarding semantic preference, we can observe expressions related to typical actions; typical places and surroundings; related professions; character traits; and appearance. Yet the corpus data also show some significant differences. The pairs of Slovenian masculine and feminine nouns referring to professions can be divided into two groups. The first one comprises pairs that are semantically equivalent and stylistically unmarked (see Derganc, 2017), while the pairs in the second group are subject to (a degree of) semantic narrowing and stylistic marking.
A prototype for the first group is the pair zdravnica/zdravnik ('female/male doctor'), both of which can co-occur with expressions referring to the type of work, typical actions and related professions. The feminine nouns in such pairs are stylistically unmarked. A similar conclusion can be reached for the pairs sodnica/sodnik ('female/male judge') and vzgojiteljica/vzgojitelj ('female/male kindergarten teacher'), but with a caveat. The form of the pair sodnica/sodnik overlap in the fields of law and sport; however, the data also reflect the societal status of the profession - the most important positions (e.g., constitutional judges) are still mostly reserved for men. The pair vzgojiteljica/vzgojitelj can be described as a borderline case between marked and unmarked meaning. The corpus data suggest a degree of semantic narrowing and stylistic marking: the meaning of "teacher in a student dormitory" refers to men, while the use of adjectives (diligent women and strict males) points to gender stereotyping.
Partially marked forms that also involve semantic narrowing have been presented using the pairs administatorka/administrator ('female/male administrator'), tajnica/tajnik ('female/male secretary'), kuharica/kuhar ('female/male cook'), maserka/maser ('masseuse/masseur'), and cistilka/cistilec ('female/male cleaner'). The masculine nouns administrator and tajnik are associated with socially more prominent professions, which is confirmed by the definitions in the SSKJ2. Administratorka tends to be in a subordinate position (notary), and is often described as shapely (appearance) or kind (character). A similar conclusion can be reached for tajnica, which co-occurs with expressions signalling her objectification (ficus plant) and sexualization (affair). Socially-constructed differences have been identified for the pair kuharica/kuhar as well. The feminine noun appears in specific work environments (parsonage) and with professions that require lower qualifications, while the masculine noun appears in
contexts related to competitions. Still, in this case, the meaning is not narrow enough to justify, for instance, separate dictionary definitions. The same can be concluded for the pair maserka/maser, which demonstrates, again, that feminatives are more likely to appear with expressions describing appearance (cute) and sexual activities (erotic). The final pair -cistilka/cistilec - shows that the profession is often associated with a specific ethnic group (Bosnian) and with stereotypes about this group (the name Fata). It can be also observed that this is a profession dominated by women, since the masculine form is mostly used to describe appliances.
6 Conclusion
The contrastive corpus analysis of collocators in English and Slovenian presented in this paper has shown that in both languages there are cases in which the feminine nouns can be described as stylistically unmarked. However, the study has identified numerous cases of masculine and feminine nouns for professions occurring in contexts that reflect the status of their referents in society. Due to these extra-linguistic factors, feminine nouns are more frequently accompanied by stereotypical views of gender roles, appearance and character: thus, their female referents are often sexualized and objectified. The masculine nouns, on the other hand, typically highlight the important position of their referents.
In the end, it should be pointed out that the main focus of the paper has been a synchronic analysis of the subject matter, relying on corpora that mostly contain examples from written language. Considering these limitations, we hope that our findings will motivate future research on the use of masculine and feminine nouns also from a diachronic perspective as well as from the point of view of different text-types.
Corpora and Tools
SSKJ2: Dictionary of Standard Slovenian, www.fran.si. BNC: https://old. sketchengine. co .uk/. EnTenTen13: https://old.sketchengine.co.uk/. FidaPLUS v2: https://old.sketchengine.co.uk/. SketchEngine, 2018: https://www.sketchengine.eu/.
References
Bartsch, Sabine. (2004). Structural and functional properties of collocations in English: a corpus study of lexical and pragmatic constraints on lexical cooccurrence. Tübingen: Gunter Narr.
Bauer, Laurie. (1996). English Word-formation. Cambridge: Cambridge University Press.
Caldas-Coulthard, Carmen Rosa, Moon, Rosamund. (2010). Curvy, hunky, kinky: Using corpora as tools for critical analysis. In Discourse & Society No 21/2. 99133. https://doi.org/10.1177/0957926509353843.
Crowther, Jonathan, Dignen, Sheila, Lea, Diana (Ed.). (2002). Oxford Collocations Dictionary for Students of English. Oxford: Oxford University Press.
Derganc, Aleksandra. (2017). Poimenovanje za zenske nosilke poklicev in polozajev v slovenscini in ruscini [Naming for women holders of professions and positions in Slovenian and Russian]. In Jezikoslovni zapiski [Linguistic notes] No 23/2. 125-130. https://doi.org/10.3986/JZ.23.2.6906.
Gablasova, Dana, Brezina, Vaclav, McEnery, Tony. (2017). Collocations in corpus-based language learning research: identifying, comparing, and interpreting the evidence. In Language Learning No 67/S1. 155-179. https://doi.org/10.1111/lang.12225.
Gabrovsek, Dusan. (2014). Extending binary collocations: (lexicographical) implications of going beyond the prototypical a - b. In ELOPE: English language overseas perspectives and enquiries No 11/2. 8-20. https://doi.org/10.4312/elope.1L2.7-20.
Gabrovsek, Dusan. (2005). Words Galore: Aspects of general and Slovenian-English contrastive lexicology. Ljubljana: FF UL, Oddelek za anglistiko in amerikanistiko.
Hoey, Michael. (2005). Lexical Priming: A New Theory of Words and Language. London: Routledge.
Holmes, Janet. (2001). A corpus-based view of gender in New Zealand. In Gender Across Languages: The linguistic representation of women and men. Marlis Hellinger and Hadumod Bussman, ed. Amsterdam: John Benjamins. 115-136.
Huddleston, Rodney, Pullum, Geoffrey K. (2002). The Cambridge Grammar of the English Language. Cambridge: Cambridge University Press.
Jurko, Primoz. (2015). Evaluative Meaning in Translation. In ELOPE: English language overseas perspectives and enquiries No 12/2. 193-209. https://doi.org/10.4312/elope.12.2.193-209.
Kilgarriff, Adam. Baisa, Vit. Busta, Jan. Jakubicek, Milos. Kovar, Vojtech. Michelfeit, Jan. Rychly, Pavel. Suchomel. Vit. (2014). The Sketch Engine: ten years on. In Lexicography No 1/1. 7-36. https://doi.org/10.1007/s40607014-0009-9.
Kranjc, Simona, Ozbot, Martina. (2013). Vloga spolno obcutljivega jezika v slovenscini, anglescini in italijanscini [The role of gender sensitive language in Slovenian, English and Italian]. In Druzbena funkcijskost jezika: vidiki, merila, opredelitve (Obdobja 32) [Social Functionality of Language: Aspects, Criteria,
Definitions]. Andreja Zele (ed.). Ljubljana: Znanstvena zalozba Filozofske fakultete. 233-239.
McIntyre, Dan. (2018). Irony and semantic prosody revisited. In The Pragmatics of Irony and Banter. Manuel Jobert and Sandrine Sorlin (eds.). Amsterdam: John Benjamins. 81-99.
Philip, Gill E. (2010). Why prosodies aren't always present: Insights into the idiom principle. In Proceedings of the Copus Lingusitics Conference CL2009. Michaela Mahlberg, Victorina González-Díaz and Catherine Smith. Liverpool: University of Liverpool. URL: http://ucrel.lancs.ac.uk/publications/cl2009/.
Quirk, Randolph et al. (1999). A Comprehensive Grammar of English. Harlow: Longman.
Rychly, Pavel. (2008). A lexicographer-friendly association score. In Proceedings of Recent Advances in Slavonic Natural Language Processing, RASLAN. Petr Sojka and Ales Horák, ed. Brno: Masaryk University. 6-9.
Sinclair, John. (1996). The search for units of meaning. In Textus: English Studies in Italy No 9. 75-106.
Stubbs, Michael. (2001). Words and Phrases. Corpus Studies of Lexical Semantics. Oxford: Blackwell.
Stubbs, Michael. (2007). Quantative data on multi-word sequences in English: The case of the word world. In Text, Discourse and Corpora: Theory and Analysis. Michael Hoey, Michaela Mahlberg, Wolfgang Teubert and Michael Stubbs, eds. London: Continuum. 163-189.
Sorli, Mojca. (2013). Corpus-based lexicographical descriptions with a special focus on pragmatics: the case of the Slovene lexical database. In ELOPE: English language overseas perspectives and enquiries No 10/2. 45-65. https://doi.org/10.4312/elope.10.2.45-65.
Sorli, Mojca. (2012). Semanticna prozodija v teoriji in praksi - korpusni pristop k proucevanju pragmaticnega pomena: primer slovenscine in anglescine [Semantic prosody in theory and practice - a corpus approach to the study of pragmatic meaning: Slovene and English]. In Dvojezicna korpusna leksikografija: slovenscina v kontrastu: novi izzivi, novi obeti [Bilingual corpus lexicography: Contrasting Slovenian: New challenges, new outlook]. Mojca Sorli (ed.). Ljubljana: Trojina, zavod za uporabno slovenistiko. 90-116.
Starnberger, Saska. (2015). Veränderungen in der gesellschaftlichen Stellung von Frauen in Slowenien seit 1991: Personenbezeichnungen im Wörterbuch Slovar novejsega besedja slovenskega jezika (2012). In New Approaches to Gender and Queer Research in Slavonic Studies. Dennis Scheller-Boltz, ed. Wiesbaden: Harrasowitz. 101-109.
Toporisic, Joze. (2000). Slovenska slovnica [Slovenian grammar]. Maribor: Obzorja.
Uhlik, Mladen. (2019). Izrazanje spolske korelacije pri polozajnih poimenovanjih v ruscini in slovenscini [Expressing gender correlation in positional denominations in Russian and Slovenian]. In Slavisticna revija [Slavistic revue] 67/2. (In press).
Vidovic-Muha, Ada. (2018). Slovensko skladenjsko besedotvorje [Slovenian syntactic word formation]. Ljubljana: Filozofska fakulteta.
