English binominal noun phrases as seen through their Czech counterparts

English binominal noun phrases as seen through their Czech counterparts Gabriela Brůhová#— Kateřina Va&ků (Charles University( Prague) ABSTRACT !is paper provides an analysis of "00 English binominal noun phrases (BNPs) and their Czech translation counterparts, using data from aparallel English-Czech corpus. BNPs, characterized by the N%-of-N& structure, have distinct syntactic and semantic properties, where N& typically functions as the semantic head, and N% conveys an attributive quality. !e study aims to identify Czech structures performing similar functions and to specify the roles of BNPs through their Czech translations. !e methodology involves manual extraction of BNPs from acorpus, classifying Czech translations into eight categories, and analyzing their evaluative, descriptive, or intensifying functions. Findings highlight the prevalence of adjectival premodification in Czech translations, the nuanced roles of BNPs, and the cultural and linguistic adaptations in translation. !e research underscores the complexity of translating expressive constructions and o(ers insights into linguistic typology and translation strategies. KEYWORDS binominal noun phrases, contrastive linguistics, expressivity, evaluative function, semantic head DOI https://doi.org/%0.%)*%&/%+0,-.",.&0&,.&.) 1 INTRODUCTION Based on the data from aparallel English-Czech corpus, the present paper o(ers an analysis of "00 English binominal noun phrases (BNPs) as seen through their Czech translation correspondences. BNPs constitute adistinctive category of syntactic constructions characterized by the combination of two nouns linked by the preposition of. !ese N%-of-N& constructions are described as “aspecial case of prepositional apposition” (Quirk et al., %-+,, p.%&+)), exemplified by phrases such as ahell of aproblem, that idiot of aprime minister, her nitwit of ahusband, the fool of apoliceman, and an angel of agirl (Aarts, %--+). BNPs are notable for their unique syntactic and semantic properties. Syntactically, according to prior research, it is the second noun that functions as the head, with the first noun and the preposition of acting as amodifier in the chunk [N% of a], e.g. hell of a(Aarts, %--+; Kim and Sells, &0%,).1 Semantically, the first noun [N%] conveys aproperty or quality attributed to the referent of the second noun [N&] (Aarts, %--+, p.%%+). !is predicative aspect of the first noun manifests itself in two main ways: the potential for paraphrasing BNPs as copular constructions (e.g., the fool of apoliceman becomes the policeman is afool) and the ability to substitute 1 The modifier construction can therefore be seen as being embedded in alarger nominal construction. OPEN ACCESS GABRIELA BRŮHOVÁ#— KATEŘINA VA,KŮ -./ [N% + of + a] with an adjective (e.g., the fool of apoliceman transforms into the foolish policeman). !is dual paraphrasing capability underscores the semantic /exibility and richness of BNPs in English. !is paper has atwofold aim: a) to identify and describe the Czech means which perform the same functions as the English BNP, b) to further specify the functions of the construction as seen through the Czech counterparts. 2 THEORETICAL BACKGROUND English binominal noun phrases exhibit arange of properties that have been the subject of linguistic analysis in works such as those by Aarts (%--+), Keizer (&00*), Kim and Sells (&0%,), and ten Wolde (&0&&). !ese properties provide insights into the structure and semantics of BNPs in English. One fundamental characteristic of English BNPs is the obligatory use of the preposition of. !is preposition is integral to the complex chunk [N% of a], which as awhole functions as amodifier (Aarts, %--+, p.%%-; ten Wolde, &0&&, p.&-%; Napoli, %-+-, p.&&)). For example, in the phrase jewel of a city in “And it introduced her to Budapest, ajewel *(of) acity.”, the omission of the preposition of would render the phrase grammatically incorrect. !e canonical determiner of the first noun in these phrases is typically an indefinite article. However, variations are possible (Keizer, &00*, p.++(), including demonstrative pronouns (that idiot of aboy), possessive pronouns (her nitwit of ahusband), and indefinite pronouns like one (one hell of abeating) or some (some dragon of areceptionist) (Kim and Sells, &0%,, p.))). !is /exibility in the choice of Det% allows for arange of nuances in BNPs. !e second determiner, on the other hand, is invariably an indefinite article, e.g., her stupid nitwit of ahusband. (Huddleston and Pullum, &00&, p.))&), or ahell of aday (Kim and Sells, &0%,, p.))–),), with other determiners not being possible (cf. ahell of *some/*any/*one day). However, in some cases, especially with plural nouns, the article may be omitted, as in jewels of villages like West Burton and Askrigg (Kim and Sells, &0%,, p.))–),). Akey aspect of BNPs is that it is the second noun (N&) that typically functions as the semantic head, contributing to the core meaning of the BNP. For instance, in She’s afrightened li"le mouse of awoman, the phrase essentially means She is afrightened woman, not She is afrightened li"le mouse. However, there are instances where both N% and N& can serve as semantic heads, as seen in She doesn’t want to talk to this idiot of aprime minister, where idiot and prime minister both hold semantic weight: She doesn’t want to talk to this idiot. or She doesn’t want to talk to aprime minister.2 (Kim and Sells, &0%,, p.),). Interestingly, the first determiner can extend its scope to encompass not only the first noun but also the second noun. !is is evident in phrases like your fool of afather (Kim and Sells, &0%,, p.)*) or our miserable sod of acleaner (Aarts, %--+, p.%"%), where 2 However, amore suitable paraphrase would contain adefinite determiner: She doesn’t want to talk to the / this prime minister in this case. OPEN ACCESS -.0 LINGUISTICA PRAGENSIA 1/1213 the evaluation applies to the person denoted by the second noun (father, cleaner): Your father is afool. or Our cleaner is amiserable sod. From the pragmatic standpoint, BNPs are described as invariably evaluative. One common characteristic of BNPs is that they often involve negative assessments of the referent of N& or are used as insults (Aarts, %--+, p.%&%). Examples like afool of adriver, abitch of asecretary, and aslut of aschoolteacher demonstrate the negative evaluative nature of BNPs. By combining N% with N&, these phrases intensify the negative judgment or derogatory nature of the description. However, it is important to note that BNPs can also be used for positive assessment when an appropriate N% is employed (Aarts, %--+, p.%&%). For instance, phrases like an angel of agirl, apearl of ajob, and awonder of acity express positive evaluations. In these cases, BNPs create asense of admiration or emphasize the exceptional qualities of the described person or thing. In terms of meaning, “there does seem to be acore of ‘expressive’ meaning in all BNP examples, and this expressive nature is presumably how the construction is functionally motivated in actual usage” (Kim and Sells, &0%,, p.,%) . BNPs o(er avaluable means to express evaluations or judgments of persons and things, particularly in fiction . !ey are typically used for the description of individuals and things to convey astrong evaluative meaning and create vivid imagery. !us, BNPs allow speakers or writers to convey their subjective evaluations with added emphasis and intensity, evoking aparticular emotional or descriptive e(ect. Den Dikken (&00., p.%.&) distinguishes two subtypes of the construction: a comparative (ajewel of avillage) and an attributive subtype (an idiot of adoctor). In summary, English BNPs are characterized by their obligatory use of the preposition of, /exibility in determiner choice, the typical role of N& as the semantic head, the potential for both N% and N& to have premodifiers, and their invariably evaluative nature.3 !ese properties contribute to the rich expressive potential of BNPs in English. 3 MATERIAL AND METHOD !e study focuses on examining English BNPs and their Czech counterparts. !is comparative analysis seeks to deepen our understanding of the structural and functional characteristics of English BNPs by exploring how they are translated into Czech, thereby shedding light on the semantics of the English construction through its Czech counterparts. Since there is no direct counterpart to the BNP in Czech, our main aim is to investigate how this specific construction is translated into Czech and how its expressivity is re/ected. !e primary methodology involves extracting examples of BNPs from aparallel English-Czech corpus and analysing their translation equivalents in Czech. !e data for this study were sourced from the InterCorp parallel English-Czech 3 For adetailed discussion of the BNP properties see Kim and Sells (2015) and ten Wolde (2022). OPEN ACCESS GABRIELA BRŮHOVÁ#— KATEŘINA VA,KŮ -.4 corpus, specifically focusing on English originals from the fiction genre. To identify English BNPs, we used the following query: [tag = "N."][word = "of"][lemma = "a"][tag = "JJ"]{0,%}[tag = "N."]. However, this approach resulted in numerous false positives, making manual selection essential for accurately identifying and extracting authentic BNPs. Consequently, we chose arandom set of authors and titles and processed the query results until we compiled "00 examples of BNPs. !e final sample contains )+ titles by ". authors. !e manual selection process for extracting BNPs was guided by two primary criteria. !e first criterion was reformulation by predication, where the BNP allows paraphrasing through acopular predication (e.g. agiant of aman translating to the man is agiant), or aparaphrase by asimile (e.g. ahoney of ahouse translating to the house is like honey). !e second criterion was that the second noun (N&) in the BNP structure should be the semantic head. !is approach, however, presented challenges, particularly with problematic examples. For instance, in the phrase shadow of a, di(erent contexts led to varying interpretations. In she was the shadow of agirl, it aligns with BNP criteria (“the girl is ashadow”), but in she saw the shadow of agirl, ambiguity arises—it could mean “she saw the girl” (BNP) or “she saw the girl’s shadow” (not aBNP). In the case of she stepped into the shadow of agirl, the structure doesn’t meet the BNP criteria. !ese nuanced contexts necessitated careful manual analysis to ensure accurate identification of genuine BNPs. !e extracted data were then classified into eight categories based on the translation counterparts in Czech. !ese categories include premodification, postmodification, clausal counterpart, synthetic counterpart, apposition, intensifier, adjectival phrase and other means. !is classification allows for adetailed analysis of how the structural and semantic properties of English BNPs are conveyed in Czech. Within each category, we determined the function the [N% of a] sequence with respect to the following noun. We identified three di(erent functions: evaluative, descriptive or intensifying. In the case of evaluative function, the [N% of a] expresses apositive or negative evaluation of N& (that fat nightmare of agirl, ahoney of ahouse). !e descriptive function is used when the [N% of a] sequence describes the size, shape or appearance of the N& referent (apink-faced, gold-haired, overgrown cherub of aman) and, lastly, the [N% of a] sequence can be used to intensify aproperty which is either expressed by amodifier of N& (ahell of astrong word “avery strong word”) or which can be inferred (hell of an accident “avery serious accident”). Our hypothesis is that the primary type of the Czech counterpart will di(er according to the function of the BNP construction. !e research is expected to provide insights into the translation strategies used for BNPs in Czech, revealing the linguistic and cultural considerations that in/uence these translations. Additionally, the research aims to deepen our understanding of the functional motivations driving the use of BNPs in English. OPEN ACCESS -32 LINGUISTICA PRAGENSIA 1/1213 4 ANALYSIS .5VARIABILITY OF N- !e sample of "00 examples of the construction contains %&% di(erent types at the N% position. !e following figure displays the most prominent types with their frequency in the corpus4: figu%e 1:Most frequent nouns in the N1 position !e dataset exhibits amarked constructional preference for the noun hell (e.g. ahell of aparty). Additionally, afairly strong preference is observed for several other nouns, such as shadow, devil, fool, ghost, giant (e.g. ashadow of afrown, adevil of asituation, that fool of aconstable, agiant of alad). However, the N% position within these constructions also exhibits ahigh degree of lexical diversity, as evidenced by the inclusion of awide array of other low-frequency types, suggesting aversatile pattern of noun incorporation. .51 VARIABILITY OF N1 !e sample contains even more di(erent types at the N& position, that is %*& di(erent nouns. Nouns with statistically significant frequency are presented in Figure &. We can reveal adiscernible pattern in the distribution of nouns occupying the N& position within our constructional framework. Unlike the N% position, which is dominated by aconspicuous outlier, hell, the N& position showcases greater variation with no single noun disproportionately frequent. !is suggests amore dynamic and /exible utilization of nouns in the N& slot. !e nouns man and time appear most frequently, but with counts of &) and &" respectively, they do not overshadow other nouns significantly. !is is indicative of aconstruction that integrates typically general nouns such as man, woman, and thing, re/ecting atendency towards general rather than specific descriptions in the N& position. 4 We applied abinomial probability model to assess the statistical significance of linguistic variant frequencies of both N1 and N2. We chose to limit our discussion to types with 4 and more occurrences, as nouns with lower frequencies displayed lower confidence interval bounds of zero, indicating insu%cient statistical evidence to conclusively di&erentiate their presence from random chance within the given confidence level (p>0.05). hell shadow devil fool ghost giant figure 5 7 8 10 15 16 110 OPEN ACCESS GABRIELA BRŮHOVÁ#— KATEŘINA VA,KŮ -3Based on the examples of the construction provided in literature we expected that most N& referents will be animate, but ..% of the dataset display an inanimate N& referent. As for the typical semantics of N&, the most frequent referents include aperson (ahawk of aman, an imperious whirlwind of awoman, ahell of aguy), time (ahell of atime, ahell of aday; anightmare of ayear), place or location (this dull hot brute shit of astreet; the tall, derelict bird’s nest of ahouse; some tourist trap of ahotel) or nouns describing human looks or characteristics (her squashed-tomato of anose; Goyle’s low rasp of avoice; abi"er tinkle of alaugh; abig crinkled Uof asmile). !e N& is often ageneral, semantically redundant noun, the main function of which is mainly to fill in the grammatical slot in the construction. .56 TRANSLATION COUNTERPARTS figu%e (:Most frequent nouns in the N2 position premodification postmodification clausal counterpart synthetic counterpart other intensifier + adv/adj apposition adjectival phrase 106 56 36 33 26 21 14 8 man time thing woman day way guy girl ways smile room place doubt 4 4 4 4 4 5 5 5 6 9 10 23 24 figu%e ):Distribution of Czech translation counterparts of the English BNP OPEN ACCESS -31 LINGUISTICA PRAGENSIA 1/1213 .565PREMODIFICATION BY AN ADJECTIVE !e most frequent counterpart of the English BNP in our study is premodification, observed in %0. instances, accounting for over athird of our data. Consistent with prior research, these Czech counterparts corroborate that N% attributes aproperty to N& (Aarts, %--+; Kim and Sells, &0%,), arole typically fulfilled by adjectival premodification in Czech. From the semantic perspective, the [N% of a] sequence primarily serves three functions: evaluative, descriptive, and intensifying. In Czech translations, these are rendered through corresponding evaluative, descriptive, or intensifying adjectives. As depicted in Figure ), the evaluative function is markedly more prevalent in this category than the descriptive and intensifying functions. figu%e 4:Functions of [N1 of a] corresponding to Czech premodification Evaluative function Both N% in the English BNP and the adjective in the Czech counterpart ascribe an evaluative property to the semantic head (N& in English, head of NP in Czech). Evaluation is afunction which is frequently described as the main function of BNPs (cf. Aarts, %--+). Examples from our dataset include: (%) a. your fool of amother její pra1t2ná matka [her foolish mother] b. her arsehole of ahusband její kreténsk3 man4el [her cretinous husband] c. areal prince of aman vá4n2 skv2l3 chlap [areally great guy] !e examples above represent purely evaluative function, involving either negative (exx. %a and %b) or positive (ex. %c) assessment of the referent of N&. Descriptive function Contrary to our initial assumptions, our data include &* instances where the BNP serves apurpose other than evaluation, with the [N% of a] sequence being more descriptive. Here, the primary role of the first noun is to describe the size, shape, or appearance of the N& referent. !erefore, we decided to categorize them separately. (&) a. agiant of aman obrovit3 mu4 [ahuge man] evaluative descriptive intensifying 59 27 20 OPEN ACCESS GABRIELA BRŮHOVÁ#— KATEŘINA VA,KŮ -36 b. aformidable barrel of aman takov3 dob5e ud2lan3, mohutn3 chlapík [awell-built, large man] c. apudding of afellow rosolovit3 chlapík [ajelly-like fellow] Intensifying function In certain instances, both the English N% and the Czech equivalent exhibit an intensifying function, amplifying aproperty associated with N&, which can be deduced from the context. (") a. one hell of anoise 1ílen3 rachot [an insane noise] b. one hell of an achievement zatracen3 úsp2ch [adamn achievement] Examples "a and "b demonstrate that the position of the first determiner typically occupied by the indefinite article can instead be filled by the indefinite pronoun one. As ten Wolde (&0&&, p.&+-) notes, this represents an emphatic use of one. Furthermore, Keizer (&00*, p.+-) highlights that one serves areinforcing role, thereby amplifying the intensity of the phrase hell of. .5651 POSTMODIFICATION !e second most frequent translation counterpart in our study is the use of aCzech postmodifying phrase or clause (%-% of sample data). Postmodification di(ers from premodification in that it allows for more expressive and verbose descriptions, which are particularly suitable for depicting the appearance of the N& referent. !e distribution of the three functions of the [N% of a] sequence is completely di(erent in this category. It seems that the evaluative function in Czech is more associated with the adjectival premodification than with postmodification. !e di(erent distribution of the descriptive and evaluative function in premodification and postmodification counterparts is statistically significant. figu%e 5:Functions of [N1 of a] corresponding to Czech postmodification Postmodification takes various forms, ranging from arelative clause (exx. )a, ,a), prepositional phrase (exx. )b, ,) or an adjectival phrase (ex. )c). In several instances the Czech translation counterparts contain ametaphor (sjestřábí tváří in ex. )b) or asimile (jako těsto in ex. )c), creating acomparison between the described person or thing and the noun used in N%. !e translation may also make use of an idiomatic exdescriptive intensifying evaluative 29 16 11 OPEN ACCESS -3. LINGUISTICA PRAGENSIA 1/1213 pression with acorresponding function, e.g. an intensifying [N% of a] is translated as an intensifying idiom into Czech (ex. .). !is translation approach appears more apt for complex semantic relationships than apremodifier, especially in the descriptive functions of N%. Descriptive function ()) a. her squashed-tomato of anose nos, kter3 p5ipomínal rozmačkané rajče [anose that looked like asquashed tomato] b. athin, shabby, ragged hawk of aman vyzábl3, o1unt2l3, zpustl3 mu4 sjest5ábí tvá5í [athin, shabby, ragged man with ahawk face] c. the fat pudding of agirl ta dívka nakynutá jako t2sto [the girl risen like dough] Evaluative function (,) a. the ruin of an old woman 4ena, zní4 zbyla jen sta5ičká troska [awoman who remained just an old wreck] b. Well, you’re one hell of adoctor! Teda vy jste opravdu doktor za v1echny prachy! [You are really adoctor for all the money! = You are aterrible doctor!] Intensifying function (.) one hell of acrack puklina jako hrom [acrack like thunder = ahuge crack ] .5656 CLAUSAL COUNTERPART Another relatively frequent means how to deal with the BNP in the Czech translation is substituting the semantically dense nominal construction by aclausal counterpart (%&% of sample data). In these instances, the translator opts for amore explicit, and often more verbose, manner of expression (exx. * a, b) to re/ect the expressivity of the original BNP. !ese cases show that Czech tends to explicitly express what English only implies, thus providing amore vivid description of the referent of N&. In many of these instances, the Czech use of alexical verb corresponds to an English construction with asemantically weak verb like have, take, or make complemented by an object (exx. + a, b). !e typological di(erence in the choice of nominal and verbal means in Czech and English was already described by Mathesius (%-*,, p.%0)), who noticed that “[i] n Czech, one almost invariably encounters apurely verbal predication, i.e. the action is expressed by the finite verb-form. !is is also quite common in English, but in addition, avery frequent type appears to be the verbo-nominal predication, i.e. OPEN ACCESS