Complex nominal groups are common in technical English (i.e. English for Specific Purposes, ESP) as they allow lexical items to be tightly packed into a clause. This leads to increased lexical density and syntactic ambiguity. In this paper we analyze (complex) nominal groups in technical English, assuming that it is not only the context and extralinguistic knowledge (i.e. shared technical background that the ESP teacher does not necessarily possess) that help solve the syntactic ambiguity, but that the structure of the nominal group, or more precisely the position of modifiers within the group can help disambiguate the meaning. Thus, modifiers standing farthest from the head have the least specifying potential and are followed by those which restrict the meaning of the entire nominal group. In the example steam reciprocating engine (vs.*reciprocating steam engine) the participle reciprocating has a more specific meaning and is thus closer to the head of the nominal group. The results show how the type of modification (linear or non-linear) can help disambiguate complex nominal groups. The main contribution of the paper is to help the ESP teacher, who is not a specialist in the field of (marine) engineering, process, understand and successfully teach complex nominal groups.

Genre implies formal and stylistic conventions of a particular text type, which inevitably affects the translation process. This „force of genre bias“ (Prieto Ramos, 2014) has been recognized by translation scholars as a relevant factor in the decision-making process, especially when it comes to the translation of legal texts. The common denominator of genre analysis and translation is the emphasis on communicative function, as the main focus of both is on the function of a particular linguistic unit in a (con)text. In this sense, one of the more important stages in translation is the process of adapting a target text to the genre conventions of the target culture, which means that a translator must be familiar with the genre conventions of the source and target cultures. In the Croatian linguistic tradition, genre has played a marginal role and has been associated with literary genres. In this research, it was assigned a central role as a yardstick against which maritime legal texts were analysed and their linguistic features defined. In this research, genre analysis was used as a methodological tool in the analysis of maritime legal texts originally written in Croatian and the translations of maritime legal texts into Croatian. In translation studies, analyses of translations into different languages have shown that translated texts differ from texts originally written in a particular language in various ways. Such an analysis of translated Croatian texts has not been done before. Therefore, the aim of the thesis is to analyse various features of this type of text, especially with regard to a possible realization of translation universals in Croatian translations. The topic of translation universals has attracted the interest of numerous scholars and has led to findings in various languages that shed light on both the translation process itself and translation strategies and procedures. However, the notion of “universal” features of translation inherent in the translation process itself has been controversial, and researchers have provided evidence for and against it, so that the term has been toned down over time to “tendencies” or “rules”. Regardless of the opposing views, the concept is still considered relevant because it provides insight into the translation process and decision-making during translation. The study in this thesis was conducted on two corpora of maritime legal texts. Specifically, one corpus was a unidirectional parallel corpus of maritime legal texts consisting of two subcorpora: maritime legal texts in English (MarLaw EN) and their translations into Croatian (MarLaw CRO). The other corpus was a comparable corpus of maritime legal texts in Croatian (MarLaw HR). The corpora were compiled using the online corpus analysis tool Sketch Engine, which immediately annotates and tags the corpora, but also contains ready- made corpora. The main goal of the research was to conduct a genre analysis of original maritime legal texts in Croatian and of translations of maritime legal texts into Croatian in order to determine the linguistic features of the genre and the manifestation of translation universals. It was assumed that translations will differ in their linguistic characteristics from texts originally written in Croatian, as research in other languages has shown that translations differ from texts originally written in that language. It was also predicted that explicitation and normalization will be the most common translation universals in translations. The first phase of the research involved benchmarking the generic features of maritime legal texts by comparing the MarLaw HR corpus with the two general Croatian language corpora available in Sketch Engine, hrWaC and Hrvatska jezična riznica (Croatian Language Repository). The hrWaC corpus is a web corpus compiled from the .hr domain, while the Croatian Language Repository is a corpus of standard Croatian language compiled from selected texts from literature, scientific publications, textbooks, online journals and newspapers, and non-fiction books. The comparison was based on quantitative indicators, in particular statistical results on lexical density, lexical richness, readability, average sentence length, and corpus results such as keywords, frequencies and distribution of parts of speech, ngrams, and collocations. To gain a deeper insight into the specific features, a quantitative analysis was conducted, which was complemented by a qualitative analysis that provided further explanations. The results showed that the MarLaw HR corpus has a lower type-token ratio, which is probably due to genre and thematic constraints, but the value of lexical density, as defined by Halliday (1985), is greater for the MarLaw HR corpus than for the general language corpora, indicating that the sentences in the specialized corpus are very dense and informative. The specialized corpus also has much longer sentences, which is consistent with the register of legal texts in general, which has a tendency to be all-inclusive and redundant (Bhatia 1996). Another feature of the legal register in maritime legal texts is nominalization, which is evident in a high noun-verb ratio, but also in quantitative corpus data, such as a larger number of nouns and adjectives compared to general language corpora. The specialized corpus also shows a larger number of nouns among the most frequent words, further confirming the tendency towards nominalization. The larger proportion of nouns in the specialized corpus contributes to information density (Biber 1989), since the focus is on the subject rather than the action, as well as to the impersonal, abstract style. The most frequent keywords in the MarLaw HR corpus, listed according to their keyness score, were divided into three main domains: legal, maritime and general lexis with technical meaning, which maps the range and scope of the genre. This analysis also revealed a larger proportion of verbal nouns compared to the two general language corpora and a larger proportion of prepositions, which is consistent with previous results. For verbs, the MarLaw HR corpus has a limited number of tenses, which is also consistent with the trends in the legal register. A distinctive feature of the genre is that present and future tenses are used to express obligation, while modal verbs are used as explicit expressions of strong obligation or strong prohibition. Adjectives and adverbs expressing obligation are also very common in the specialized corpus. Verbs are often decomposed into constructions consisting of a periphrastic verb and a noun, which also contributes to nominalization. Another feature of the specialized corpus is a larger proportion of performative verbs used as secondary means of expressing the imperative. At the discourse level, the specialized corpus shows an increased proportion of relative and demonstrative pronouns and certain groups of discourse markers, such as conditional, causal, relative, and explanatory. Another distinctive feature of the genre is multinomial expressions, particularly those involving verbs. The following part of the research involved the comparison of the specialized corpus of Croatian maritime legal texts with the parallel corpus of English and translated Croatian texts in order to analyse the specific features of translated texts and potential manifestations of translation universals. The analysis, also conducted in Sketch Engine, included statistical and corpus results, similar to the previous phase. Each identified feature was then associated with a particular translation universal. The statistical analysis showed that translations have the lowest average sentence length, lexical richness, and readability score, indicating their simplification compared to source and original texts. The translations also showed a similar tendency to nominalization as the original texts and a similar distribution of tenses. However, the translations also showed some peculiarities, such as lower valency of periphrastic verbs, a lower use of performative verbs, an increased use of modal verbs, and, at the discourse level, an increased use of pronouns, prepositions, and connectives. Manifestations of translation universals was considered from two points of view, that of the source text and that of the original text, in order to have a better insight into their realization. In this sense, explicitation was identified in the proportion and distribution of adjectives, pronouns, conjunctions and textual connectors, in expressing modality. The tendency to normalization was noted in the proportion and distribution of nouns and the reduction of English binomial expressions with synonymous components, the latter not being a feature of Croatian legal register. Another interesting feature is the tendency towards implicitation in the translations in the category of discourse markers, which are usually an example of explicitation in translations in other languages. Genre has been shown to be a decisive factor in identifying translation universals, as the texts written in a particular genre can be counterposed against texts written in the same genre, otherwise the results would be biased by specific generic features. It should be remembered that the specialized corpus MarLaw is limited both thematically and genre-wise, so the results obtained are specific to this particular genre. Nevertheless, it has revealed the tendencies in the translation of maritime legal texts, which can be reproduced for comparative purposes with other legal genres, and other genres to identify the differences and similarities in the translation process. The research has also contributed to the ongoing discussion on the translation universals from the perspective of the Croatian language, which has not been studied before. The results may have implications for professional translation, as well as for translator training. The research results may stimulate further research in the area of genre and translation analysis using corpus tools, as well as the compilation of other useful corpora of Croatian language to further develop Croatian language resources.

The aim of the 4 th International e-Conference on Studies in Humanities and Social Sciences (4IeCSHSS) was to bring together scholars, administrators and students from different countries, and to discuss theoretical and practical issues in different areas of Humanities and Social Sciences. The e-Conference was organized as a kind of a multi-disciplinary forum which provided the appropriate opportunities for inter-disciplinary communications.

Translation research focuses mainly on parallel and comparable corpora, whereby it is constantly faced with issues of representativeness, balance and comparability as its main constraints. This research aims to introduce the concept of genre as a way of observing linguistic features under controlled conditions. The study analyses the application of external and internal criteria with particular focus on the genre criterion in selecting texts for the compilation of a highly-specialized bilingual maritime legal corpus, consisting of source texts in English and their translations into Croatian. The main advantages and constraints of genre as a criterion are discussed. The main benefits of such an approach are found in its application in translator training and practice. In addition, genre-based approaches to corpus analysis may raise awareness of generic features specific to a target language, ultimately improving the quality of translation.