Iberica 13 Ibérica 25 (2013): 85-106 ISSN 1139-7241 Abstract In this paper we deal with some novel blends in English and discuss, from a cognitivist point of view, the ways they are formed and processed, particularly focusing on the interpretation of their meaning and the degree of recognisability of the source words in these blends by the ESP students of the Faculty of Economics, University of Belgrade, as well as the difficulties they face in understanding them. We point out various reasons why these blends tend to be misinterpreted by non-native speakers of English and discuss the extent to which ESP teachers may rely on the tenets of Conceptual Blending Theory in the process of economic vocabulary acquisition and learning in an ESP economics course at the tertiary level. Keywords: blends, Conceptual Blending Theory, understanding blends, ESP, non-native speakers. Resumen S ob re “ s il ic o na ir e s ” y “ mi l l io ne r ds ” – Có mo e n ti e nd e n l os a l umn os d e I F E l os nu e v os c ru c e s e n in g l é s En este trabajo nos centramos en diversos cruces nuevos (novel blends) que se registran en inglés y estudiamos, desde un punto de vista cognitivo, sus modos de formación y procesamiento. Concretamente nos centramos en un grupo de alumnos de IFE en la Facultad de Económicas de la Universidad de Belgrado y cómo interpretan los significados, el grado de reconocimiento de las palabras fuente que dan lugar a estos cruces y las dificultades que pueden encontrar estos alumnos en su comprensión. Señalamos diversas razones por las que los hablantes no nativos de inglés pueden llegar a hacer interpretaciones erróneas de Of “siliconaires” and “millionerds” – How ESP learners understand novel blends in English Nadežda Silaški and Tatjana Đurović University of Belgrade (Serbia) silaskin@sbb.rs & tdjurovic@sbb.rs 85 Ibérica 25 (2013): 85-106 N. SILAšKI & T. -DUrOvIĆ estos cruces y estudiamos hasta qué punto los profesores de IFE pueden adherirse a los postulados de la Conceptual Blending Theory en el proceso de adquisición y aprendizaje del vocabulario económico presente en los cursos de inglés para ciencias económicas que se imparten en la educación superior. Palabras clave: cruces, Teoría de la integración conceptual, comprensión de los cruces, inglés para ciencias económicas, hablantes no nativos. 1. Introduction Although a few blends in English were recorded as early as in the 15th century (Lehrer, 2007), blending1 has only recently become one of the most popular word formation processes and an important source of neologisms. At the same time, “[i]n spite of a recent surge of interest in it, blending remains among the most poorly understood and elusive word formation processes” (Brdar-Szabó & Brdar, 2008: 171). There are many things about blending that “remain to be investigated and so many issues on which there is no real consensus among researchers, from a satisfying definition to an extensional account of the phenomenon including an inventory of various subtypes” (Brdar-Szabó & Brdar, 2008: 172). Nevertheless, blending has hitherto received insufficient scholarly attention and has been regarded as a marginal word formation process. It is only recently, however, with the increasing attention given to Cognitive Linguistics, and Conceptual Blending Theory in particular, that blends have begun to attract the interest of researchers mostly due to “their creative nature and unusualness” (Kemmer, 2003: 76), as well as their extreme suitability for analysis from a cognitive standpoint. In addition to already lexicalised and conventionalised blends such as “smog”, “workaholic” or “brunch”, where “some speakers are no longer aware of their underlying complex sources” (Lehrer, 2007: 115), we are witnessing (almost on a daily basis) the emergence of new blends in English, many of which are one-off, nonce word forms, where speakers and hearers need to expend a great deal of effort in recognising the source words. These blends “are coined by speakers online to fit a specific communicative purpose, and may or may not become part of the shared lexicon of the linguistic community” (Benczes, 2009: 49). In this paper we deal with some novel blends in English and discuss, from a cognitivist point of view, the ways they are formed and processed, 86 particularly focusing on the disambiguation of their meaning and the degree of recognisability of the source words in these blends by the ESP students of the Faculty of Economics, Belgrade University, as well as the problems they encounter in the process of understanding these blends. On the basis of the results obtained from a purposely designed questionnaire, we point out various reasons why these blends tend to be misinterpreted by non- native speakers of English and discuss the extent to which ESP teachers may rely on the tenets of Conceptual Blending Theory in the process of economic vocabulary acquisition and learning in a tertiary-level ESP economics course. 2. The definition of blending, conceptual blending and lexical blends Blending as a word formation process “tends to shade off into compounding, neo-classical compounding, affixation, clipping and acronyming” (Bauer, 1983: 26), which is why it cannot be defined in an easy and clear-cut way. Kelly (1998: 579) defines blends as words “formed by snipping components from existing words and stitching the components together either through simple concatenation or through concatenation coupled with overlap of shared phonological segments”. Blending, on the other hand, may be defined as “the intentional coinage of a new word by fusing parts of at least two source words” (Gries, 2004: 416). The part of word which forms a blend is called a “splinter”, which cannot occur alone as a word, although it is basically a clipping. For example, in “infotainment” (information + entertainment), info is a clipping and can be used as an independent word. However, “-tainment” is a splinter and cannot stand alone but must be attached to something else (Lehrer, 2007). Kemmer (2003: 92) claims that the analysis of lexical blends under the theoretical wing of Conceptual Blending Theory “fits very well into the general theory of blending developed by Turner and Fauconnier (1995)”, since this theory views them as “just one type of blending, in which form happens to be blended as well as concepts.” Conceptual Blending Theory, as originally devised by Fauconnier and Turner (2002), builds on the notion of “mental spaces” (Fauconnier, 1994). Unlike Conceptual metaphor Theory (Lakoff & Johnson, 1980), in which a claim is made that the source domain structures the target domain (which means that the process of OF SILICONAIrES AND mILLIONErDS Ibérica 25 (2013): 85-106 87 metaphorisation involves two conceptual domains), according to Conceptual Blending Theory a third space is created, which is a separate, blended mental space.2 mental spaces are “small conceptual packets constructed as we think and talk, for purposes of local understanding and action” (Fauconnier & Turner, 2002: 40). Blending is the combination of two inputs in mental spaces that yield a third mental space called a blend, which “is not merely a composition of the first two but instead has emergent structure of its own” (Fauconnier & Turner, 1999: 76). Conceptual blending involves three basic processes – “composition”, “completion”, and “elaboration”, of which composition, “the most straightforward process, refers to the projection of content from each of the inputs into the blended space” (Grady, Oakley & Coulson, 1999) and is central to the problem dealt with in this paper. According to Ungerer and Schmid (2006: 268), “the most obvious candidate for an analysis in terms of conceptual blending is its namesake in the area of word-formation, the morphological blend, as represented by items such as smog, brunch, motel, infotainment”. However, all these blends have already become deeply entrenched and lexicalised, insomuch that “many language users will not even realise the blending background any longer” (Ungerer & Schmid, 2006: 268) and are able to understand these blends without “unpacking” them – that is without being aware of their input spaces. In this paper, however, we deal with novel blends, whose processing and understanding requires mental effort expended in their unpacking and recognising the source parts. The claim that blends usually arise due to the principle of language economy and efficiency is somewhat contradictory with the claim some authors make, namely that “most new blends and other trendy neologisms (...) don’t increase efficiency” (Lehrer, 2003: 369) but, on the contrary, “create more effort to interpret – at least at first, until readers and hearers have figured out what the source words are and what they mean” (Lehrer, 2003: 369). This is the case with novel, creative blends which, according to Ungerer and Schmid (2006: 268), are “the real testing ground for a conceptual blending analysis”, since many of them are “intentionally conceived as a temporary and open- ended phenomenon” (Ungerer & Schmid, 2006: 268) and are thus very poor candidates for conceptual entrenchment. In the next section we deal with the results obtained from a questionnaire designed to test the ability of non- native speakers of English (students of Economics) to understand and correctly interpret selected novel blends. N. SILAšKI & T. -DUrOvIĆ Ibérica 25 (2013): 85-10688 3. Materials and participants A total of 95 students, aged 21-23, took part in the survey. All students were attending their third year of studies at the Faculty of Economics at the time the survey was carried out, after having learnt General English for 12 years in their primary and secondary schools and having attended a highly specialised two-semester ESP economics course at the Faculty. The students were informed that the questionnaire involved the identification and interpretation of novel blends in English. They were also given an explanation of what blends are (the definition of blends was illustrated by the use of the example of “infotainment”, which turned out to be familiar to a vast majority of students). They were presented with a questionnaire with a list of 33 blends (mostly collected from Investopedia.com site in 2011, filed under the heading Buzz Words) divided into five segments. Each of the first four segments dealt with the blends triggered by the same word, while the last segment contained miscellaneous blends not resting on any specific trigger word.3 The students were asked: (1) to identify the source words of the blends, and (2) to interpret and explain the meaning of each blend in Serbian, since we did not want their knowledge of English, which is somewhere between B1 and B2 levels, to hinder the correct explanation of the way they understand blends from the questionnaire. Although previous research (see Lehrer, 2003; for Serbian see Halupka- rešetar & Lalić-Krstin, 2009 & 2012) indicates that blends are processed somewhat more easily in context than without context, the blends in the questionnaire were presented to the students in a decontextualised manner. This is because, as Lehrer (2003: 372) puts it, “[t]he creators [of blends], often journalists and advertisers, use these terms without definitions, expecting readers and hearers to ‘get’ them”. At the same time, however, “the problem of identification and interpretation [of blends] is highly determined” (Lehrer, 2003: 370) if blends are presented in context. Since our main goal was to check the potential of on-line processing of novel blends (as propounded by Conceptual Blending theorists) with ESP learners in order to determine the level of their understanding of novel blends and to see if non-native speakers of English approach the processing of blend meaning in the same way as native speakers do, we did not want the context to influence and facilitate their interpretation so the blends were presented in the questionnaire in an isolated way. In pursuit of investigating how students comprehend and interpret blends in a decontextualised setting, we deem that possible incorrect inferences and OF SILICONAIrES AND mILLIONErDS Ibérica 25 (2013): 85-106 89 faulty interpretations “may provide us with as much insight as correct choices.” (Charteris-Black, 1998: 15). In the following sections we discuss the main results of the survey, concentrating individually on the three groups of blends contained in the questionnaire. What follows is not a statistical but a detailed qualitative analysis of the questionnaire results, since we believe that such an analysis may allow for a richer and more profound explanation of the results which may partially or entirely be lacking if the results were analysed quantitatively, as that would not be sufficient to account for the “reasons” for possible misinterpretations of meaning. 3.1. The case of “economics” and “–nomics” blends Lehrer (2007: 120) claims that “once a blend is created, the splinter may be reused”. However, when the splinter becomes a common part of numerous blends, it may tend to lose its original connection with the source word and “can be considered as a morpheme in its own right” (Lehrer, 2007: 121), going through the process of being a completely novel splinter at first to finally becoming a completely conventional morpheme. This is most probably the case with “–nomics”, which is increasingly becoming a clear candidate for morpheme status. However, “–nomics” does not originate from a once novel blend and has never been a morpheme in its own right. Nevertheless, the “–nomics” splinter, produced by structural resegmentation of the word “economics” and thus being “endowed with some semantic autonomy” (Frath, 2005: 104), has been used to coin tens and tens of neologisms. The splinter retains the meaning of the word it replaces, although it does not exist in unbound form – for instance there is no such thing as a ratti (a splinter in glitteratti, coined according to literatti) (Frath, 2005). Similarly, the word nomics does not exist but is obtained in the reanalysis of “economics”, albeit not semantically grounded, thus producing an affix, “–nomics”, which has now acquired a bound morpheme status (Lehrer, 2007), functioning in the blends as a final splinter. As far as the degree of students’ understanding of “–nomics” blends4 is concerned, there are some general conclusions to be drawn. Firstly, the meaning of those “–nomics” blends whose source word is a personal name was most easily interpreted. This is a distinctive group of neologisms which can be defined collectively as terms used to refer to the overall economic policies of certain presidents or governments, such as “obamanomics”, N. SILAšKI & T. -DUrOvIĆ Ibérica 25 (2013): 85-10690 “nixonomics”, “clintonomics”, “reaganomics” tested in our questionnaire, and they are all metonymically based. Despite a high degree of students’ comprehension of these blends, there are certain misinterpretations which may be attributed to a number of different reasons. Firstly, several students processed the meaning of blends exclusively at orthographic/phonological level (the so-called bottom-up processing, which is based on someone’s linguistic knowledge), wrong guesses arising from misspelled splinters, such as “economic development aimed at clients” in “clintonomics”, or “regional economics” in “reaganomics”. Secondly, although “nixonomics” has produced a significant number of correct interpretations, several students misinterpreted it either as “nix” + “economics” and explained it as “without economics”, which, irrespective of its semantic content, is reminiscent of one underlying characteristic of blending – fuzzy morpheme boundaries, when source words are sometimes linked in a very surprising manner. The meaning of “Enronomics” was also readily recognised by a vast majority of students. Even the explanation that the blend refers to a “fake (falsified) economic prosperity” may fare at least as a partial understanding. Similarly to some previous blends, several misinterpretations of “Enronomics” arise from wrongly linking the first source word to the noun “environment” (thus, “economics of the environment” or “environmental economics”) in one case, or to the adjective “enormous”, thus “Enronomics” = “enormous development” in the second. This implies that although students make wrong recognition of source words at the intralingual level of graphology, they still strive and manage to make possible conceptual connections between these wrongly identified source words, which is suggestive of a high potential of language creativity of blending. “Kremlinomics”, on the other hand, is one of the rare examples of the “–nomics” blends (together with “boomernomics” and “perkonomics” that we will mention later) whose meaning has proven to be very difficult to decode, producing not one correct interpretation whatsoever. Although the students had a vague notion about the given blend due to a successful identification of the source words, they were unaware of the extra information content of the word “Kremlin”, which attests to the fact that the blended word is not a simple compounding of (mostly) two words’ meaning, but is enriched with a broader conceptual content. Only one OF SILICONAIrES AND mILLIONErDS Ibérica 25 (2013): 85-106 91 explanation, “economics of self-centeredness”, remotely recalls the true meaning of “kremlinomics”. In this particular case, therefore, the blending process is in effect entirely lacking. Some other “–nomics” blends, however, turned out to be misinterpreted by almost all students, “boomernomics” being a case in point. Out of 95 students who filled out the questionnaire, only three of them interpreted the blend and its underlying source words correctly. As many as 23 students stated that the second source word, in addition to “economics”, was “boomerang”, which of course distorted the meaning of the blend. Hence “economic boomerang”, “measures of economic policy which backfire”, “economy which causes a negative feedback”, etc. A number of students, on the other hand, stated that the meaning of “boomernomics” was “economic boom”, their interpretation of the blend resting on the understanding that “boomer” in “boomernomics” was somehow linked with economic boom periods (therefore, “economies that are progressing fast”, “a sudden change”, “economy linked up with a huge expansion”, “economic prosperity”, etc.). The students were obviously not familiar with the concept of “baby-boomers”. Although both problems that the students encountered in their interpretations may be readily attributed to the absence of sufficient knowledge about a (baby) boomer, there are rather plausible explanations for the failure to decode the meaning of “boomernomics” in both cases. The former – that is “boomernomics” as a wrongly identified blend from “boomerang” and “economics”, may be attributed to a consistent application of one of the combining patterns, that of two splinters when the fore part of the first source word is concatenated with the hind part of the second source word, with no overlapping (similarly to “brunch”). Due to the students’ failure to notice that the first source is a whole word, not a clipped part, the difficulties arise as to “the identification of underlying associations between elements” (Charteris-Black, 1998: 24), which in turn leads to the above mentioned comprehension difficulties. In the latter case, on the other hand, students fail to utilise a grammar schema based on the rule noun + the agent suffix (here, “–er”), identifying only part of the source word and trying to establish relations between apparently semantically compatible words (“boom” and “economics”), which may lend some feeble grounds for justifying the selected choice of meaning. The correct decoding of “boomernomics” is obviously governed by coherence and the application of the so-called top-down model of interpretation in global context, which N. SILAšKI & T. -DUrOvIĆ Ibérica 25 (2013): 85-10692 encompasses one’s world knowledge. The lack of culture-specific knowledge of the word “boomer” has rendered the proposed meanings of the blend “boomernomics” incorrect. Another example of poor understanding of “–nomics” blends is “burgernomics”. Namely, a number of students stated that “burgernomics” meant either “policy of fast-food restaurants” or “fast-food business”, unaware of the existence of the Big mac Index, which may be attributed to a gap in their subject-matter knowledge. However, some interpretations are evidence of interesting strategies of meaning processing, based primarily on the first source word. Hence explanations that “burgernomics” is “poor quality economy”, “cheap economy”, “unhealthy, bad development”, or “something that is composed of several parts” hinge on common-sense beliefs about the quality of fast food, or as the last example shows, on visualisation converted into a verbal form, which consequently guide interpretations of the given blend and its disambiguation. The suggested answers are a key to the students’ conceptual packing of two input spaces, even though one of them has not been fully developed due to some extralinguistic reasons. “Flexinomics” and “perkonomics” seem to be the most confusing blends to our economics students. most of them recognised the source words (“flexible” + “economics”) and (“perk” + “economics”) respectively, but were not able to get their meanings. most of them stated that “flexinomics” was a kind of “flexible economics” (or, “a high level of economic liberalisation”, liberalisation obviously understood as flexibility), but were not able to elaborate on this interpretation. “Perkonomics” was most frequently interpreted as “economics based on employees’ perks”, whatever that might mean. The problem that students encounter here lies in semantic extension, which requires searching for deeper semantic associations between the source words that go beyond their apparently transparent meaning. This is in line with Lehrer’s (2007: 117) argument that “[a]fter the source words have been identified, a plausible meaning must still be found” (Lehrer, 2007: 117). The contextual usage of these two blends would undoubtedly enhance both comprehension and better retention, bearing in mind that particularly in the example of “perkonomics” students are aware of the great semantic plausibility between the two source words. OF SILICONAIrES AND mILLIONErDS Ibérica 25 (2013): 85-106 93 3.2. The case of “millionaire” and “–naire” blends The second group contained those blends coined according to the trigger word “millionaire”: “optionaire”, “dellionaire”, “siliconaire”, “spillionaire”, and “millionerd”. The students were rather successful in finding the source words, except for “spillionaire”, where they misinterpreted the word “spill”, originally meant to refer to the British Petroleum oil spill. Namely, most students stated that a “spillionaire” was “someone who is so rich that they spill money around them”, whereas the intended meaning of this novel blend is “someone who makes millions due to the British Petroleum oil spill in the gulf ”. A completely opposite explanation that it is “someone who has lost – that is, spilled – a million” also adds to a problem of exclusion of some meanings of source words, in this case “spill”, which tends to be indispensable so as to comprehend the meaning of this blend altogether. In other words, different interpretations may be here ascribed to “a partial and selective transfer of meanings” (Charteris-Black, 1998: 26; original bold), additionally coupled with the absence of any contextual cues. “Siliconaire” turned out to be the most interesting case of “millionaire” blends. Thus, a vast majority of students made their own on-line meaning construction, interpreting the meaning of “siliconaire” as “a plastic surgeon who makes millions on silicone breast implants” or, conversely, “the plastic surgeons’ big salaries” themselves. Obviously unaware of the different spelling of “silicon” (a chemical element) and confusing it with “silicone” (rubber-like material used for breast implants), students formed and selected their own conceptual packets. In addition, several explanations draw on what Charteris-Black (1998: 28) labels as “syntactic opacity”, when students are not able to restore the missing syntactic elements so as to paraphrase the blend as intended. This is evidenced in interpretations that “siliconaires” are “silicones for millionaires”, “people who have so much money that they can waste it on silicones”, or even that these are “millionaires with silicone implants” and consequently not real but “fake millionaires”! According to Charteris-Black (1998: 11), foreign students “may lack culture-specific knowledge to provide the semantic basis for an interpretation which can enable them to supply deleted syntax”. Another wrong but perhaps culturally-dependent interpretation, was that a “siliconaire” was “a woman, who, thanks to her breasts augmented with silicone implants, married a millionaire” and a variation of it that it was “someone who makes millions working in show business as singers, actors/actresses”, or even more “informed” interpretations that it was “someone who has made a fortune N. SILAšKI & T. -DUrOvIĆ Ibérica 25 (2013): 85-10694 thanks to some artificiality”, “an artificial millionaire, who assumes an air of being a millionaire, not being actually one”, “new money”, or “a millionaire who has suddenly become one”! The source words of both “optionaire” and “dellionaire” were rather easily recognised, but the meaning processing was still hard to perform. Thus, several students stated that an “optionaire” was “a millionaire who can afford to keep his options open, who has many options in life”, or “the one who uses their options” and “has the option of becoming a millionaire”. Few interpretations were wrong due to semantic reasons since “optionaire” was said to be “an optional millionaire”, “optional” interpreted as “potential”5, whereas in some cases students incorrectly identified the first source word as “optimisation”, hence “optionaire” turned out to be “a millionaire by way of optimisation”, or it meant “to optimise with the aim of becoming a millionaire”. Wrong source word recognition was also evident in an explanation in which it was stated that an “optionaire” was “a millionaire who has made a fortune from optical business”. “Dellionaire” also proved to be semantically opaque, giving rise to only a few correct meaning construals. misinterpretations may be generally attributed to a poor source word identification as evidenced in one student’s explanation as “a not so rich millionaire, something like Del boy from Only fools and horses”6, and several related ones as “funny (millionaire)”, “not so well-off a millionaire”, or “someone who has become rich in retailing”. Wrong disambiguation as “a millionaire in delirium” also shows the same type of comprehension problem (together with the explanation “decision making”?!) where the blend was split in a totally unexpected way. misspelling again proved to be a rich source of erroneous answers, a full word “dell” mistakenly decoded as “deal”, but being in fact (phonologically) related to Serbian verb deliti “to divide” or “to share” on the one hand, which is why “dellionaire” was understood as a process of “dividing/sharing wealth”, “dividing millions”, or to deliti “to deal playing cards among players” on the other, expanding the meaning of the blend “dellionaire” to “a millionaire who has something to do with betting”. Finally, “millionerd” (“millionaire” + “nerd”) originally meaning “a wealthy person who made their money in computer software or some other high- tech industry”, was correctly interpreted by most students, which probably arises from their familiarity with the source word “nerd” as well as its frequency. Still, some interpretations bear witness to the fact that the word OF SILICONAIrES AND mILLIONErDS Ibérica 25 (2013): 85-106 95 “nerd” has frequently been associated with attributes deemed as pejorative, such as being too preoccupied with studying and void of social ability. This conceptual content found its way in explanations such as “a nerd who has made millions”, “a nerd who has become rich”, “someone obsessed with making money”, “a stingy person”, “a nerd – bourgeois”, or a very offensive “a retarded millionaire”. However, several examples show that faulty interpretations of the students echo certain grammar rules, producing in turn interesting answers. Thus, in explanations “ex-millionaire”, “having become a millionaire”, “a person worth a million or more dinars or some other currency”, or “valued at million of something” the blend meaning seems to be deducted from a morphological process of participle-adjective conversion, when any past participle may be used as an adjective. This indicates that blend comprehension as a kind of lexical inferencing of meaning is a very complex process with foreign learners and rests on all available linguistic cues and knowledge as well as the learners’ mental lexicon which enables them to make many possible semantic associations, irrespective of their conceptual correctness or intention. A significantly higher number of students who successfully decoded the meaning of “millionerd” confirms Lehrer’s (2003: 372) hypothesis that “[b]lends in which the targets are frequent would be processed more quickly than those which are less frequent.” In addition, we may assume that a high degree of the students’ exposure to this blend (via the use of the Internet) has also to be accounted for. 3.3. Miscellaneous blends The third group of blends from the questionnaire contained miscellaneous blends (“funemployment”, “diworsification”, “returnment”, “homepreneur” and “blamestorming”) although not selected randomly. The main criterion for the blend selection was their morphological composition which we tried to make as versatile as possible. Thus, the first blend was “funemployment” whose composition may be difficult to establish: it may be understood either as “fun” + “unemployment” (zero splinter plus partial overlap) or, alternatively, “fun” + “employment”, in which case the understanding of the blend would not be right. As many as 66 of the total number of 95 students stated in the questionnaire that the source words of this blend were actually “fun” + “employment”. Consequently, they wrongly interpreted the meaning of this blend as “a job which is fun to do”, “funny employment”, “a job for fun”. Alternatively, due to syntactic problems, students also N. SILAšKI & T. -DUrOvIĆ Ibérica 25 (2013): 85-10696 explained that the given blend meant “employment in the amusement/entertainment industry” or “employment in show business”. Similar to some previous blends, a wrong identification of source words may be attributed to phonological/orthographic similarity which may still produce semantically viable interpretations. Thus in one case a student managed to identify “unemployment” as the second source word, but considered “fu–” a splinter of the word “full”, hence “funemployment” resulted in “full unemployment”. In two additional cases, erroneous interpretations were accounted for by a poor identification of both source words, giving rise to “fundamental employment” in one case (“fun[damental]” + “employment”), or “functional employment” (“fun[ctional]” + “employment”) in another. This once again points out all the intricacies of blending in spite of the students’ awareness of the general principles of melding two words together and is in line with Kemmer’s (2003: 77) argument that “[r]ecognizability is a matter of degree (…) and does not lend itself to formulation in a rule”. It generally shows that blends with partial or complete overlap are very difficult to process, in the sense that they may allow for different source identification depending on how many neighbouring words can be derived (as the “funemployment” example shows – that is, “fun” can be “fun(ny)”, “fundamental”, “functional”). As Lehrer (2003: 372) states “[b]lends without neighbors or with less frequent neighbors will be processed faster than those with more frequent neighbors”. Let us add here that sometimes even the aspect of semantic plausibility may be seriously questionable – one may agree that “unemployed individuals who decide to enjoy the free time that unemployment provides” is equally semantically permissible as “full employment” or “functional employment”, as proposed by some students. It implies that blending asks for digging deeper for conceptual combinations between source words, which may sometimes be considerably helped by contextual cues. The second blend in the miscellaneous section of the questionnaire was “diworsification” (“diversification” + “worse”, in which there is an embedded splinter). From the point of view of Serbian students, phonologically as well as graphically speaking, this blend seemed to resemble the word “divorce”, giving rise to a host of amusing but completely wrong explanations, such as “legal divorce” (where “–fication splinter was identified as “nostrification” [sic!]), “a document on divorce”, “suffer because you have recently divorced”, or even “a divorce bringing pleasure”. A wrong recognition of source words gives reasons for interpreting “diworsification” OF SILICONAIrES AND mILLIONErDS Ibérica 25 (2013): 85-106 97 as “to split vacation, not to use it immediately” (“divorce” + “vacation”) or “a classification after dividing up of a company” (“divorce” + “classification”). Even though we recorded three correct explanations of the given blend, a good deal of either incorrect answers or no answers at all account for the finding that, regarding the ease of identification, embedded splinter counts as the most difficult (see Lehrer, 2003). The next blend “returnment”, whose effectiveness lies in phonological similarity between the two source words, “return” and “retirement”, was correctly decomposed by only a few students. A large number of suggested wrong interpretations revealed that the students focused their attention only on the word “return” in order to match Serbian translation of this verb with the words it usually collocates, practically disregarding the second splinter, and thus failing to hint at the right meaning of the blend. Hence misinterpretations range from “return (reimburse) of money or salary”, “to get (return) a job back”, “to return (pay) a debt by paying out”, “hiring/firing”, additionally explained as “return employment”, “return of invested assets”, “return on investment”, to a surprising and different one “relocating the employees to other work places”. These processing flaws indicate that in certain cases students of foreign language resort to their native language for semantic clues, with a smaller or higher degree of interference. Out of all the blends in the miscellaneous section, “homepreneur” has shown to be correctly identified and comprehended by the majority of students, which may be attributed to their high familiarity with the technical word “entrepreneur” and even more, with a high frequency word “home”. Even though they certainly recognised the source word “home”, some students experienced problems in trying to establish target semantic and conceptual links between the two inputs. Accordingly, proposed glosses, such as “someone who performs house jobs”, “family business”, “a self- made entrepreneur”, “a sole employer”, “a small entrepreneur”, “a domestic entrepreneur”, “someone who runs a household”, “a housekeeper”, “the self-employed”, or even “a person/job of decorating the interior of a home” show on the one hand, that some students process blends as if they were noun compounds, and on the other, that the students attempt to decode the meaning by forming a vast array of semantic associations in their semantic networks triggered in this case by the word “home”, although they are rendered incorrect as explanations of “homepreneur”. misspelling was again the reason for one conceptually illogical explanation, in which “home” was N. SILAšKI & T. -DUrOvIĆ Ibérica 25 (2013): 85-10698 taken to be “homo” and the whole combination understood as “the same gender employee”. Some of the above examples also show syntactic deficiencies, when students do not discriminate between someone who works “from” home and someone who works “at” home. Thus, one student showed this difference by explaining that “homepreneur” is “a prisoner”, combining conceptual and phonological aspects in a very surprising way indeed. Finally, the blend “blamestorming”, which rests on a phonological familiarity with the word “brainstorming” that projects a greater part of its meaning onto “blamestorming”, proved extremely demanding with regard to its successful comprehension, due to the wrong identification of input elements and demonstrating a complete disregard for the otherwise familiar word “brainstorming”. Since the majority of students simply separated the blend into “blame” and “storm(ing)”, they missed the underlying semantic and conceptual links between the two source words and produced interpretations which rely either on the first or on the second element. Hence “a rush of blame”, “gathering collective blame”, “guilt”, “an attack of blame”, “a set of blame one after another”, or simply and only “storm”. 4. Discussion Novel lexical blends, mostly realised as a concoction of usually well-known source words, operate at a deeper conceptual level rather than at a superficial linguistic level unlike noun compounds to which they are very similar, so “the tight lexical integration of two distinct word-forms into a unified lexical whole suggests an equally tight integration of ideas at the conceptual level” (veale & Butnariu, 2010: 403). According to the questionnaire results, a good deal of the ESP students of the Faculty of Economics, Belgrade University, are faced with the very problem of successfully projecting the meaning of input words and its subsequent integration into a blend. Some explanations provided by the students regarding blends such as “blamestorming” and “kremlinomics” exemplify the problem of successful conceptual coalescing of the two source words even when the students have successfully identified them. Their usage in a sentential context would perhaps facilitate the proper interpretation. However, we tested the students’ ability to recognise and interpret blends in a decontextualised setting, in real time, without being backed by some background knowledge cues. The way some students OF SILICONAIrES AND mILLIONErDS Ibérica 25 (2013): 85-106 99 decompose blends and construct their meaning, although in an incorrect or unexpected way, has provided us with several insights concerning on-line blend processing and its potential as a tool of vocabulary learning of ESP economics students. Generally speaking, the working out of blend meaning resorts to accessing words from the available students’ mental lexicon trading on the information which stands at their disposal, and that is closely connected with the overall foreign language proficiency of the students. In other words, we may conclude that the higher the level of English knowledge, the higher the degree of successful blend interpretation. Blend comprehension, where students perform lexical inferencing tasks, should account for “making informed guesses as to the meaning of a word in light of all available linguistic cues in combination with the learner’s general knowledge of the world, her awareness of context and her relevant linguistic knowledge” (Haastrup, 1987: 197). On the one hand, this means that the students, although making wrong guesses, deploy the so-called top-down processing when a blend is viewed as a “semantic package” which contains all the information that a student can relate to identifying source words. This is evident in explanations that “clintonomics” is “economics of being wasteful”, “Enronomics” is “a fake economic prosperity”, “obamanomics” is “economics of saving”, etc. This may also suggest that in some cases the students view blends as gestalts – that is, the processes of identification and interpretation of source words are probably simultaneous or the process of interpretation is at least performed with no considerable mental effort in comparison to the one of identification. On the other hand, instantiations of bottom-up processing are evident in those (incorrect) explanations where students experience problems concerning syntactic relations between source words. Thus, for example, the question which arises is the following: is a “siliconaire” “someone who is a millionaire because of the business of implanting silicones”, or “a millionaire with implanted silicones”? The second major finding relates to the problem of meaning disambiguation when, despite the correct identification of source words, the students still produce wrong guesses, due to word polysemy. The blend “siliconaire” is a case in point, where the students focus on a meaning of “silicon/e” different from the one intended. In spite of an attempt to locate the right meaning of a word in its semantic network some students simply miss the semantic coherence between the source words, which consequently causes their wrong inference on a conceptual level as well. As some experiments show (Lehrer, N. SILAšKI & T. -DUrOvIĆ Ibérica 25 (2013): 85-106100 1996 & 2007), these are the factors that should be accounted for with native speakers as well. Alternatively, other (especially metaphorical) meanings of words may not be simply overlooked but not be part of the students’ vocabulary at all, in which case the intended meaning would be entirely lacking. The third finding is linked up with one of the most important problems in blend comprehension, the problem emerging at a phonological level, since phonological properties are extremely relevant in blending and the wrong identification of blend source words results in semantic and conceptual incomprehension. Incorrect processing of “returnment” or “diworsification” by some students underlines the aspect of phonological similarities of input words and their effect on blends. Furthermore, as Kemmer (2003: 77) says “[t]he amount of similar structure can vary a great deal, so it is impossible to state a general formal rule that will license some blends and exclude others”, which proves to be an obstacle to a successful blend processing with both native speakers and foreign learners. If the previously mentioned findings account for parallels that can be drawn between native speakers of English and non-native ESP learners regarding the construction of the blend meaning, then problems associated with the processing of, for instance, “burgernomics”, “boomernomics”, “optionaire”, “dellionaire”, or “spillionaire” pinpoint the significance of general background knowledge of the student, including his/her subject- specific knowledge. Although this kind of information is beyond the realm of linguistic knowledge, it is frequently the crucial carrier of the blend meaning. If the students had been aware of the extralinguistic meaning of the input word “boomer” in “boomernomics”, for example, they would have more successfully integrated it in the given blend. Context, frequency of usage, and exposure to blends can probably compensate for the missing cultural and subject-specific links, indispensable to a higher level of the students’ successful inferencing of the blend meaning. 5. Pedagogical implications Our findings reveal significant pedagogical implications for the teaching of novel blends in an ESP classroom at tertiary level. Namely, they clearly indicate that English teachers cannot expect that their students, non-native speakers of English, will be able to process the meaning of novel blends on- line in the same way native speakers of English will, that is novel blends OF SILICONAIrES AND mILLIONErDS Ibérica 25 (2013): 85-106 101 “have to be taught”, just like other units of vocabulary. Among the above mentioned reasons for the misinterpretation and misunderstanding of novel blends, the most important in our opinion is that many cultural gaps, necessary to be filled before the process of meaning construction even starts, prevent the students from getting the right meaning of blends, while at the same time, their own cultural milieu often impedes the correct interpretation of blends. The correct “unpacking” of novel blends by non- native speakers of English requires the right cultural background, and if the knowledge of that background is missing, that is if extralinguistic opacity occurs, novel blends remain either incomprehensible or misdecoded. Therefore, the apparent lack of conceptual fluency, characteristic of learners of English as a foreign language, is due to the fact that they think in terms of their native conceptual system (Yu, 2009: 299). moreover, “the various pieces of information that are associated with any given concept are largely socially constructed” (Littlemore, 2004: 269) and “a given item is likely to activate a different set of associations” (Littlemore, 2004: 275), since the blending process and particularly the formation of conceptual packets, in Conceptual Blending Theory terms, is a highly individualised act which cannot guarantee the grasping of intended meaning of blends. 6. Conclusion In this paper an attempt has been made to account for the results of a questionnaire which tested the students’ understanding of novel blends in English in terms of Conceptual Blending Theory. As far as teaching novel blends is concerned, we may conclude by saying that conceptual blending, as it is claimed to be naturally occurring in native English speakers, cannot be a reliable ground for the teaching and the adoption of novel blends’ meaning and their supposed on-line meaning construction when it comes to ESP learners. many preconditions mentioned above need to be met before non- native and native speakers’ conceptual packets match and before they select the same conceptual content which will be cognitively blended into the same verbal manifestation. As far as suggestions for further study are concerned, it would be interesting to conduct some research among ESP learners to analyse the extent to which factors such as the context in which blends appear, knowledge of English, general education, or any other relevant variable help in the “unpacking” and understanding of blends. It would also be interesting for future research to compare the understanding of the same N. SILAšKI & T. -DUrOvIĆ Ibérica 25 (2013): 85-106102 novel blends by the native speakers of some other European languages to establish the similarities or differences between the ways they combine and blend the available conceptual packets and the ways this is done by Serbian speakers. Acknowledgements The paper is the result of research conducted within project no. 178002 Languages and cultures across space and time funded by the ministry of Education, Science and Technological Development of the republic of Serbia. [Paper received 23 April 2012] [Revised paper accepted 13 October 2012] References OF SILICONAIrES AND mILLIONErDS Ibérica 25 (2013): 85-106 103 Bauer, L. (1983). English Word-formation. Cambridge University Press: Cambridge. Benczes, R. (2009). “What motivates the production and use of metaphorical and metonymical compounds?” in M. Brdar, M. Omazić & V. Pavičić Takač (eds.), Cognitive Approaches to English: Fundamental, Methodological, Interdisciplinary and Applied Aspects, 49-69. Newcastle upon Tyne: Cambridge Scholars. Brdar-Szabó, R. & M. Brdar (2008). “On the marginality of lexical blending”. Jezikoslovlje 9: 171-194. Bugarski, R. (2001). “Dve reči u jednoj: leksičke skrivalice”. Jezik danas 13: 1-5. Charteris-Black, J. (1998). “Compound nouns and the acquisition of English neologisms”. URL: http://www.eric.ed.gov/PDFS/ED427525.pdf [10/05/2011] Fauconnier, G. (1994). Mental Spaces: Aspects of Meaning Construction in Natural Language. Cambridge: Cambridge University Press. Fauconnier, G. & M. Turner (1999). “Metonymy and conceptual integration” in K-U. Panther & G. Radden (eds.), Metonymy in Language and Thought, 77-90. Amsterdam/Philadelphia: John Benjamins. Fauconnier, G. & M. Turner (2002). The Way We Think: Conceptual Blending and the Mind’s Hidden Complexities. New York: Basic Books. Fischer, R. (1996). Lexical Change in Present-day English: A Corpus-based Study of the Motivation, Institutionalization, and Productivity of Creative Neologisms. Tübingen: Gunter Narr Verlag. Frath, P. (2005). “Why is there no ham in hamburger? A study of lexical blends and reanalysed morphemisation”. RANAM (Recherches Anglaises et Nord-Américaines) 38: 99-112. Grady, J.E., T. Oakley & S. Coulson (1999). “Blending and metaphor”. URL: http://markturner.org/blendaphor.html [30/04/2009] Gries, S.T. (2004). “Isn’t that fantabulous? How similarity motivates intentional morphological blends in English” in M. Achard & S. Kemmer (eds.), Language, Culture, and Mind, 415-428. Stanford: CSLI Publications. Haastrup, K. (1987). “Using thinking aloud and retrospection to uncover learners’ lexical inferencing procedures” in C. Faerch & G. Kasper (eds.), Introspection in Second Language Research, 197-212. Clevedon: Multilingual Matters. Halupka-Rešetar, S. & G. Lalić-Krstin (2009). “New blends in Serbian: typological and headedness-related issues”. Godišnjak Filozofskog fakulteta u Novom Sadu XXXIV: 115- 124. Halupka-Rešetar, S. & G. Lalić-Krstin (2012). “Razumevanje slivenica u srpskom jeziku”. Nasleđe 22: 101-110 Nadežda Silaški works as an Associate Professor of English for Economists at the Faculty of Economics, University of Belgrade. She got her PhD degree in linguistics in 2005. Her fields of interest include discourse analysis, cognitive linguistics and ESP. She is the author of a monograph on Anglicisms in Serbian economic register and co-author of a monograph on metaphors in Serbian public discourse as well as author and co-author of several ESP economics textbooks. Tatjana Đurović works as an Associate Professor of English for Economists at the Faculty of Economics, University of Belgrade. She received her PhD in linguistics in 2006. Her main research interests include cognitive linguistics, ESP and discourse analysis. She is the co-author of a monograph on metaphors in Serbian public discourse and author and co- author of several ESP economics textbooks. NoTes 1 Blending has now been widely accepted as a term. However, blends have been named differently in linguistic research (for example “portmanteau words”, “contamination”, “telescope words”, “amalgams”, “brunch-words”, “compromot”, among others). See Fischer (1996) for a detailed discussion. In Serbian, Bugarski (2001) proposes the term slivenica (“blend”), which is also accepted by Klajn (2002) and Halupka-Rešetar and Lalić-Krstin (2009 & 2012). N. SILAšKI & T. -DUrOvIĆ Ibérica 25 (2013): 85-106104 Kelly, M.H. (1998). “To “brunch” or to “brench”: Some aspects of blend structure”. Linguistics 36: 579-90. Kemmer, S. (2003). “Schemas and lexical blends” in H. Cuyckens, T. Berg, R. Dirven & K.-U. Panther (eds.), Motivation in Language: Studies in Honor of Günter Radden, 69-97. Amsterdam: John Benjamins. Klajn, I. (2002). Tvorba reči u savremenom srpskom jeziku I. Novi Sad: Institut za srpski jezik i Matica srpska. Lakoff, G. & M. Johnson (1980). Metaphors we Live by. Chicago: Chicago University Press. Lehrer, A. (1996). “Identifying and interpreting blends: an experimental approach”. Cognitive Linguistics 7: 359-390. Lehrer, A. (2003). “Understanding trendy neologisms”. Rivista di Linguistica 15: 369-382. Lehrer, A. (2007). “Blendalicious” in J. Munet (ed.), Lexical Creativity, Texts and Contexts, 115-133. Amsterdam/Philadelphia: John Benjamins. Littlemore, J. (2004). “What kind of training is required to help language students use metaphor- based strategies to work out the meaning of new vocabulary”. D.E.L.T.A. 20: 265-279. Turner, M. & G. Fauconnier (1995). “Conceptual integration and formal expression”. Metaphor and Symbolic Activity 10: 183-203. Ungerer, F. & H.-J. Schmid (1996/2006). An Introduction to Cognitive Linguistics. London/New York: Longman. Veale, T. & C. Butnariu (2010). “Harvesting and understanding on-line neologisms” in A. Onysko & S. Michel (eds.), Cognitive Perspectives on Word Formation, 399-421. Berlin/New York: Walter de Gruyter. Yu, N. (2009). From Body to Meaning in Culture. Amsterdam/Philadelphia: John Benjamins. 2 For a detailed account of the relationship between Conceptual Metaphor Theory and Conceptual Integration Theory see Grady, Oakley and Coulson (1999). 3 The blends were activated by the words “economics”, “millionaire”, “downsizing” and “vacation”, but only the results pertaining to the first two trigger words, “economics” and “millionaire”, will be discussed in this paper. This is due to a space constraint, but much more importantly to the fact that the students’ strategies of blend recognition and understanding tended to repeat so that the first two groups of blends, triggered by “economics” and “millionaire”, will suffice in illustrating and explaining those strategies. 4 “Boomernomics”, “burgernomics”, “clintonomics”, “Enronomics”, “flexinomics”, “kremlinomics”, “nixonomics”, “obamanomics”, “perkonomics”, and “reaganomics”. 5 There is a clear mother tongue transfer involved here, since the Serbian word opcioni “optional” means, among other things, “alternative” or “possible”, being similar both orthographically and semantically to potencijalni “potential”. 6 A character from the famous British sitcom. OF SILICONAIrES AND mILLIONErDS Ibérica 25 (2013): 85-106 105