The Language of Alternative Lifestyles A Critical Analysis of the Discourses of Emos and LOHAS Bernhard Kettemann & Georg Marko This article examines how two alternative lifestyles - LOHAS (= Lifestyle of Health and Sustainability) and Emos (= youth lifestyle stressing negative emotionality and social withdrawal) - try to construct their Otherness, i.e. their deviance from a postulated mainstream, in their discourses. Starting from a Critical Discourse Analytical perspective, we analyse two small corpora representing the respective discourses, trying to demonstrate in the process how modern means of computer-assisted and quantitatively oriented corpus analysis can fruitfully be applied to the investigation of cultural issues. Sample analyses of word lists, keywords and the collocations of select linguistic structures reveal that Emos define their status as outsiders less via reference to a mainstream culture but rather by solipsistically focusing on their inner and emotional selves. LOHAS, on the other hand, seems to be pushing a reformist agenda, concentrating on our - human and non-human nature - collective well-being. This agenda, however, seems to be defined as achievable through economic measures, which creates the impression that LOHAS has partly become part of the capitalist mainstream culture rather than being a real alternative. 1. Introduction Social identities are produced by the continuous process of interpretation rather than being objectively given. Self-identification therefore primarily happens in discourse and thus in and through the incessant play with signifiers and signifieds. Structuralism assumes that elements in a sign system are mainly defined in relation to other elements, which means their identity rests on what they are not as much as on what they are. Transferred to the field of society and culture, this means that our identities - if we think of them as AAA - Arbeiten aus Anglistik und Amerikanistik Band 37 (2012) · Heft 1 Gunter Narr Verlag Tübingen Bernhard Kettemann & Georg Marko 70 values in a system - could also derive from distancing ourselves from a certain group, i.e. from actively not being part of it. Constructing an alternative identity (in the sense of ‘different from others’, not ‘different from our own ‘standard’ identities’) thus is a process of undermining, reversing, or perverting the signifiers of mainstream culture. As mainstream culture, however, is not a given but rather a construct, people adopting such alternative identities may unintentionally contribute to a solidification of this construct. Or in other words, by presenting themselves as different and as transgressing the borders of conventions, they reconfirm the existence of these very conventions and strengthen their relevance. In this article, we want to examine how two alternative groups - Emos and LOHAS supporters - engage in this active Otherness and differentness. We will be paying more attention to their language than to their use of other signifiers (e.g. food, clothes). Our objective is to demonstrate how such an analysis of discourse can be done and how we can shed some light on processes of identity construction rather than to present the results of a comprehensive and coherent study. 2. What are alternative lifestyles? Before dealing with analyses proper, we want to examine the concept of alternative lifestyle more closely. We will additionally provide some background information on the two groups we will be concerned with. 2.1. Lifestyle The title features the word lifestyles rather than groups, culture, or identity. So the first question to ask is: What is the difference? Lifestyle has become something of a buzzword in current social theory, partly because some proclaim lifestyles to be a phenomenon of modern society cutting across and partly replacing more traditional categories such as class, gender or ethnicity (cf. Chaney 1996). A lifestyle can be defined as any open set of freely-chosen practices and items by which I express and make perceptible who I am or who I want to be in terms of ‣ Beliefs (epistemological systems) ‣ Attitudes (systems of ethical values) ‣ Tastes (system of aesthetic values) They thus define my social identity, by which I position myself in a society (cf. Marko 2010: 148f.). The distinctive features of lifestyles are the following (adopted and adapted from Chaney 1996): The Language of Alternative Lifestyles 71 I. The symbolic value of everyday life This is the idea that in today’s Western society, things are not just what they are but they also are signs of something else. These things potentially include everything, i.e. what you wear, what you like, what you eat, what you do, etc. Another interesting aspect relevant in modernity is that there is a shift - obviously promoted by advertising and the mass media - from functional to cultural meanings. Products of everyday life used to indicate - and thus to mean in a broad sense - their function: a door meant ‘you can enter/ leave here’, stairs meant ‘you can walk up/ down here’, a pepper mill meant ‘you can grind pepper with this’, etc. Advertising, of course, primarily referred to the functional qualities of a product, in the sense of ‘this is a good pepper mill because it grinds pepper so much better than anything else’. Today, we buy a pepper mill because of its nice design (meaning: we have a good taste), because of the higher price (meaning: we can afford it or, more likely, we spend money on these kind of things because these details are important to us), because of the brand and its connotations (one of which might, interestingly, be that it covers highquality products), etc. Signifiers in late modernity and particularly regarding the phenomenon of lifestyles are primarily visual. In the era of the anonymous city, the spectacle, i.e. things to see, have become even more central than before. It is not by chance that public conceptions of a person are referred to as their image. The importance of visuality also explains the salience of the body - its shaping and trimming through food, exercise and clothing - in lifestyles. II. The predominance of choice In today’s Western societies, people are constantly confronted with choices which are no longer determined by institutions, but which - to a certain extent - have to be taken individually. And with choices we create meanings, which in turn contribute to constructing our social identities. Consumption is central in these choices, more so than production. Lifestyles are consequently primarily defined by signifiers from the area of a hedonistic form of consumption. Even though lifestyles appear to be individual choices, they rely on social categories and social meanings to work. III. The centrality of social/ personal identity The main meaning/ signified of cultural signifiers in modernity concern the question of who I am, i.e. my social identity. In the structuralist/ poststructuralist tradition, it is assumed that there is no ontological found- Bernhard Kettemann & Georg Marko 72 ation to identity, i.e. there is no real me out/ in there, but it is an interpretation of the signifiers that I use. So who I am is equivalent to who I signify/ represent to be. My identity is no longer - at least not to the same extent - determined by my social class and my working position, my geographical origin, my gender or my ethnicity. Any of these may become dominant, but only to the extent that I choose to. But more commonly, identities are fragmented, with different aspects playing a significant role in different contexts or with aspects mixing, partly in heterogeneous combinations. In lifestyles, identities are first and foremost defined ethically and aesthetically. This means who I am is primarily defined by the things that I consider good and bad and by the things that I consider beautiful and ugly and by my acting accordingly. 2.2 Lifestyle communities People who share a lifestyle may form a coherent interactive community, which means they maintain some kind of contact with others that is motivated by and based on this lifestyle. In this case, we speak of a lifestyle community. The interactive contact, however, is not a necessary prerequisite to a common lifestyle. This implies that lifestyle identities can be more privatized than primarily class-based identities, even though any kind of lifestyle needs the public that can read the signifiers pointing to a certain identity. 2.3 Alternative If lifestyles depend on individual choices, we may wonder whether there could be such a thing as an alternative lifestyle - or a mainstream lifestyle, for that matter - shared by a large group of people. But the question is not about ‘reality’, but about interpretations and perceptions (even though we could probably argue that there are certain statistical trends in people’s choices which allow them to subsume them under a common denominator). Alternative then means ‘in contrast to a perceived and postulated mainstream lifestyle and its concomitant conventionality’. This contrast is pursued actively, so that we can say that alternative lifestyles actively seek to position themselves as the Other, with the implication that this is also the Better. We can distinguish three types of alternative lifestyles: ‣ Escapist: People adopting escapist lifestyles distance themselves from mainstream culture, seek as little contact as possible with the latter and show no interest in changing it either. Example: selfsupporting communities. The Language of Alternative Lifestyles 73 ‣ Confrontational: People adopting confrontational lifestyles actively and possibly even aggressively confront mainstream culture, either because they think that conflict provides some legitimation for their own lifestyles or because they seek a revolutionary upheaval of mainstream culture. Example: punks. ‣ Reformist: People adopting reformist lifestyles seek to reform mainstream culture gradually by education. Their own lifestyles serve as models to be emulated by others. Example: some forms of vegetarianism. The ideal 1 objective shared by all types is to disclose the problematic nature of mainstream lifestyles. The question, however, is whether they - perhaps to a different extent - actually sustain and legitimate these very lifestyles, perhaps also because alternative lifestyles of any kind are socially dependent on mainstream lifestyles (punks would lose their special outlaw status if everybody adopted their ideals, vegetarians would not be super-ethical if no one ate meat, etc.). The other interesting question is to what extent mainstream culture manages to appropriate alternative lifestyles by adopting their signifiers (jeans with cuts and holes, organic food supermarkets, etc.) (cf. Fiske 1994). 3. Two examples: Emos and LOHAS We have decided to focus on two lifestyles, namely Emos and LOHAS (= Lifestyle Of Health And Sustainability). Even though they clearly classify as alternative, they represent two different poles of the spectrum. They therefore seem to suggest themselves for a comparison. 3.1 Emo Emo - a clipping of emotional - is the label of an alternative youth culture characterised by introversion and withdrawal from an outside (adult) world perceived as unsympathetic, misunderstanding and imposing and the concomitant emphasis on negative and depressive moods, (self-) humiliation, self-pity and suicidal ideas. Externally, Emos show a preference for dark colours in clothes, hairstyle and make-up and for androgynous styles. 2 Being an Emo seems to be a phenomenon strictly limited to teenagers. 1 Ideal because it is far from clear whether practicing Otherness remains on the level of the signifiers or whether this corresponds to some deeper levels of meaning. 2 The term was initially used for a genre of music blending hardcore punk with emotionality. Bernhard Kettemann & Georg Marko 74 Emos keep in touch with each other, mostly through computer-mediated communication. We might even argue that Emos construct and enact their identities primarily in and through this medium, with localized Emo identities, i.e. identities based on the interaction with one’s immediate social environment, not playing an important role. Emos are difficult to subsume under the three categories posited above. But they probably fit best into the escapist category because they do not seem to be looking for changes in mainstream culture. They, however, maintain some kind of contact with the latter as a source of humiliation, from which they derive some justification for their identities (cf. Kelley/ Leslie 2007). 3.2 LOHAS LOHAS is an acronym standing for “Lifestyle Of Health And Sustainability”. It refers to a lifestyle that combines a focus on a positive promotion of (personal) health and well-being with environmentalist attitudes. Signifiers of a LOHAS lifestyle range from the practicing of Yoga or Qi Gong and visits to spas and other wellness locations via an openness towards alternative and complementary medical procedures (acupuncture, Ayurveda, etc.) to buying organically grown and fairly traded food and using environmentally friendly forms of housing and transportation. Those practicing LOHAS belong to a well-educated and well-to-do segment of the population. It is not clear to what extent people would explicitly define themselves as LOHAS (the term itself tends to be used to characterize a group from a marketing point of view). And those subscribing to LOHAS clearly do not - in toto - form lifestyle communities. LOHAS can be categorized as reformist alternative because most supporters believe that theirs is a way of life to be followed. This, of course, does not preclude the possibility that some enjoy considering themselves elite. But rigid elitism does not agree with LOHAS principles (cf. Kirig/ Wenzel 2009 and LOHAS online). 3.3 Why should we study Emos and LOHAS? Alternative lifestyles can be regarded as a critique of mainstream culture. Studying them provides an opportunity to bring to light what those explicitly and intentionally positioning themselves outside the centre of a society think about the latter and about problematic aspects of mainstream culture and conventionality. Such research also allows us to consider the nature of the alternative paths that such - in a neutral sense - deviant lifestyles offer, their consistency or - possibly - their inconsistencies, contradictions and their dependencies on the very mainstream at which they target their critique. The Language of Alternative Lifestyles 75 LOHAS and Emo appear to represent very different forms of alternative lifestyles. Studying them may show two distinct approaches to Otherness and distancing from mainstream, both with their peculiar problems. This makes a comparison all the more interesting. 3.4 How should we study Emos and LOHAS? Since the linguistic turn in the social sciences, there has been agreement that language does not only play a crucial but an even constitutive role in social, political, cultural and historical phenomena. As these phenomena are based on shared conceptions and attitudes and as the process of sharing happens in and through interaction, examining language use and communication appears to be a viable and valuable path to understanding.We will therefore focus on the language that is used by supporters of the two alternative lifestyles under scrutiny. 4. Methodology and data 4.1 Approach We are taking a Critical Discourse Analysis (or CDA) approach. CDA aims to elucidate/ understand sociocultural and sociopolitical phenomena (usually in contested areas) by examining the role that language plays in these. It assumes that how language is used regularly (in particular discourses) contributes to the creation and maintenance of certain systems of beliefs and attitudes (= ideologies). This of course also applies to ideologies underlying alternative lifestyles. Subscribing to a functional paradigm, CDA assigns great importance to even small linguistic elements, which might have large-scale effects on meanings and interpretations. Critical Discourse Analysis thus examines the nitty-gritty details of authentic language use, describing patterns of forms that can be related to the interpretation of patterns of meaning which in turn may correspond to patterns of conceptualizations whose socio-cultural significance (usually in terms of ideology) we evaluate (for more thorough introductions to CDA, cf. Marko 2008, Fairclough 1989, 1992). 4.2 Method Critical Discourse Analysis is an approach rather than a method, since the three procedures -description, interpretation, evaluation - just mentioned could be carried out by various means. This said, it normally centres on a Bernhard Kettemann & Georg Marko 76 semantically-oriented form of textual analysis. But even within this, methodology can be differentiated. There is no one-to-one relationship between the levels just mentioned and sociocultural significance can therefore not be read off from linguistic forms. But consistencies in form and meaning across a large number of different texts may add weight to our conclusion that particular meanings play a role in people’s conception of a certain domain and may thereby gain socio-political significance. We are not interested in the extraordinary and the unique, but in the ordinary and in typical patterns. This suggests that a purely qualitative study of individual pieces of discourse might not suffice  but we need larger samples of text. This is why we choose corpus analysis, i.e. the computer-assisted examination of large electronically-stored collections of text, as our method in CDA. It allows us to retrieve patterns, interpret them in their verbal contexts, and quantify them (for an introduction to the use of corpora in all forms of discourse analysis, cf. Baker 2006). 4.3 Data We will be using two relatively small corpora representing the discourses used by LOHAS supporters and Emos. In both cases, finding suitable material proved more problematic than initially assumed. The Emo corpus was compiled by Kerstin Florian for a paper in a seminar entitled “The language of alternative lifestyles”. She included a heterogeneous set of genres written and made publicly available by young people identifying as Emos, such as blogs, fashion and lifestyle articles, poems and song lyrics. Corpus linguists call such collections “quick and dirty” (cf. Tribble 1997). But it can be argued that the corpus in its inconsistent composition represents the chaotic textual universe in which young people shape and enact their Emo identities. 3 The corpus comprises 141,614 word tokens. The LOHAS corpus was compiled by ourselves for this paper. We suggested above that LOHAS has not yet gained widespread acceptance as a label. So people clearly falling into this group might not explicitly identify as LOHAS. This, however, makes it difficult to decide what to include and what not. We therefore chose to limit ourselves to a website explicitly associating with LOHAS (LOHAS online at http: / / www.lohas.com) and to copy all articles posted or included on this site (24 overall). It has to be mentioned that the material does not always clearly distinguish between LOHAS support or LOHAS marketing and promotional interests. But probably this is a typical feature of LOHAS discourse. 3 Kettemann 2011 presents results from the same Emo corpus in a similar fashion as we do here but from a pedagogical perspective, discussing how corpus data can help in the teaching of Cultural Studies. The Language of Alternative Lifestyles 77 The corpus comprises 27,507 word tokens. Both corpora are relatively small in size. But there is a strand in corpus linguistics that assumes that small corpora, if focusing on very specific discourses or genres, can be relevant sources of information (cf. Tribble 1997). We have to admit, though, that for a more comprehensive study corpus sizes would have to be increased and the sizes should be adjusted. For the analysis, Wordsmith Tools 5.0 by Mike Scott (2008f.) was used. 5. Exploring the discourses of Emos and LOHAS We will proceed along a quasi narrative path. This means we will demonstrate how we normally start exploring the data to arrive at the most significant and interesting issues. We will partly do this comparing the two discourses in question, partly examining a particular question only with respect to one of them. 5.1 Wordlists Corpus-based discourse analysis tends to focus on words and word patterns as the constituents of meanings across individual texts. This is why starting with a look at words promises first insights into the conceptual grid which discourses construct. Concordancing programmes of the quality of WordSmith have a function that can list all the words - defined here as orthographic words, i.e. as combinations of letters - found in a corpus with their frequencies. A raw frequency list for the two corpora looks like this. Emo LOHAS I 4,312 the 1,363 the 3,206 and 983 to 2,834 of 793 and 2,649 to 773 you 2,553 in 529 2007 2,220 a 528 my 2,197 is 420 a 2,036 that 326 by 1,720 for 293 of 1,462 are 264 me 1,432 as 206 it 1,424 s 186 pm 1,360 with 180 is 1,343 LOHAS 159 12 1,286 on 145 Bernhard Kettemann & Georg Marko 78 in 1,152 more 141 that 1,143 by 140 Emo 1,102 it 133 am 961 organic 113 but 922 their 113 so 842 from 112 for 824 green 110 your 813 has 106 1 761 products 100 with 730 or 94 Table 1: Wordlists of the Emo corpus and the LOHAS corpus. These lists already show interesting differences. The fact that the first person pronoun I tops the list, with second person you also in the top 5 in the Emo corpus stresses the rather conversational tone of most of the texts and also the personal nature of the content. On the LOHAS side, we already see some terms that seem significant with respect to the content, e.g. LOHAS, organic, green, products. Overall, however, the problem with such lists is that they contain lots of grammatical words such as articles, pronouns or prepositions, which contribute to meanings, but usually only in combinations with lexical words. We have therefore produced lists filtering out these grammatical elements (and technical elements referring to dates and times of posts in blogs included in the Emo corpus). Refined in this way, the lists look as follows. Emo LOHAS Emo 1,102 LOHAS 159 like 721 organic 113 just 657 green 110 hair 530 products 100 love 472 consumers 85 need 355 energy 81 know 338 business 79 life 326 companies 67 think 319 says 67 help 294 growth 65 post 287 environmental 62 joined 275 consumer 61 posts 275 new 58 heart 268 yoga 57 people 263 market 55 posted 254 percent 55 want 247 world 54 see 242 health 52 The Language of Alternative Lifestyles 79 really 239 sustainable 50 top 235 years 50 current 234 spa 49 feel 234 company 48 subject 230 clean 46 pain 222 technology 43 black 220 use 42 say 220 medicine 41 look 218 people 41 music 212 food 39 location 211 industry 39 make 209 association 38 Table 2: Content wordlists of the Emo corpus and the LOHAS corpus. The most frequent words of these edited lists indicate that Emos live between the existential sphere of emotions and thinking and their fashion, while LOHAS adds a strong economic slant to their focus on health and ecology. To examine these aspects in more depth, we have taken the first 250 lexemes from these lists (the full lists, this is, not the extracts shown here) and categorized them according to their semantic domains. For this purpose, we have combined grammatical realizations of the same word, e.g. singular and plural or present and past tense, and partly also morphologically related words, e.g. sustainable and sustainability, under single headings (technically speaking, we have lemmatized the corpus, cf. Mukherjee 2009: 67). As there is no pre-given and fixed set of semantic domains, we put together a list based on previous attempts at semantic categorization (cf. Archer/ Wilson/ Rayson 2002), on our expectations concerning the themes deal with in the texts, and on informal explorations of the data. A subjective and arbitrary element can, however, not be fully avoided. We used the following domains: ‣ Existence & change: processes of being and becoming; e.g. exist, become. ‣ Persons/ humans: e.g. man, Jane, priest. ‣ Location: e.g. inside (as an adverb), place. ‣ Time: concepts concerned with or related to points in time, duration, age, frequency, etc.; e.g. old, today, century. ‣ Possession: concepts concerned with or related to having, getting or losing; e.g. own, obtain, lose. ‣ Phase: concepts concerned with or related to starting, finishing or continuing; e.g. stop, begin, keep on. Bernhard Kettemann & Georg Marko 80 ‣ Modality: concepts concerned with or related to likelihood; e.g. likely, perhaps. ‣ Evaluation: attributes expressing a judgement.  Positive: e.g. great, good, top.  Negative: miserable, weak, stupid.  Comparative: similar, different. ‣ Quantity: concepts concerned with or related to amounts, size and numbers; e.g. big, amount, growth. ‣ Colour: e.g. black, pink. ‣ Body: concepts concerned with or related to the body and its parts; e.g. head, face, hair, body. ‣ Health: concepts concerned with or related to the preservation of health and the treatment of ailments and diseases; e.g. spa, massage, cure. ‣ Social phenomena:  Social groups: e.g. punks, group, Emos.  Social institutions: (or related to them); e.g. organization, run (an organization).  Social relations: e.g. mother, parents, bisexual. ‣ Communication: acts, products and media of communication; e.g. say, talk, book. ‣ Mental phenomena:  Cognition: thinking; e.g. know, remember.  Perception: sensual perception; e.g. see, hear, look at.  Emotion: affection, volition and evaluation; e.g. fear, enjoy, need. ‣ Science: concepts concerned with or related to research, its administration and presentation; e.g. research, study, example. ‣ Geography/ politics: concepts concerned with or related to nations or countries; e.g. America, local, home. ‣ Chemical substances: e.g. antioxidant, carbon. ‣ Ecology: concepts concerned with or related to nature and its protection and to natural phenomena; e.g. renewable, solar, wind, nature. ‣ Economy: concepts concerned with or related to the economy, business and industry; e.g. bank, business, economic, trade. ‣ Music: concepts concerned with or related to music and the music business; e.g. music, rock, bands ‣ Style: concepts concerned with or related to clothes, make-up and hairstyle; e.g. pants, wear, dye. The Language of Alternative Lifestyles 81 ‣ Computer: concepts concerned with or related to computer technology and the world wide web; e.g. internet, gallery. ‣ Physical contact: concepts concerned with or related to physical contact with people or objects; e.g. touch, cut, kiss. We then calculated the sizes of the different categories with respect to both lexical variation (i.e. how many different words are included in a category) and token frequencies (i.e. how often do all the words in a category occur). Both can be argued to contribute to the salience of a particular domain. If there are a lot of synonymous, near-synonymous or at least semantically closely related words - a phenomenon referred to as overwording (cf. Fairclough 1989: 115, Goatly 2000: 64) - then this creates the impression that a discourse is preoccupied with this domain. Similarly, if words occur very frequently, then this also foregrounds the specific domain. The following table thus contains the relative sizes of the most important categories with respect to both lexical variation and token frequencies. The percentages represent the relative sizes of the classes relative to the overall number of words. A percentage of 25% thus means that a fourth of all words found in among the top 250 content words belong to the respective class. Mind that for the calculation of percentages, unclassifiable words were also considered. The totals therefore do not add up to the sum of the individual categories. Emo LOHAS Existence & change 2 1.0% 736 3.1% 5 1.9% 102 1.5% People 8 4.1% 945 4.0% 9 3.4% 191 2.9% Location 9 4.6% 866 3.6% — — — — Time 14 7.1% 1330 5.6% 11 4.2% 294 4.5% Possession 3 1.5% 291 1.2% 2 0.8% 29 0.4% Phase 3 1.5% 274 1.2% — — — — Modality 3 1.5% 367 1.5% — — — — Evaluation 21 10.7% 1,914 8.0% 28 10.1% 577 8.9% Quantity 2 1.0% 199 0.8% 22 8.3% 481 7.4% Colour 5 2.5% 488 2.0% — — — — Body 7 3.6% 1,108 4.7% 1 0.4% 11 0.2% Health — — — — 8 3.0% 300 4.6% Social 17 8.6% 3,132 13.1% 26 9.8% 692 10.7% Group 3 1.5% 1,314 5.5% 6 2.2% 302 4.7% Institution 1 0.5% 82 0.3% 11 4.2% 226 3.5% Relation 7 3.6% 763 3.2% 2 0.8% 25 0.4% Communication 12 6.1% 1,330 5.6% 4 1.5% 125 1.9% Music 3 1.5% 306 1.3% — — — — Style 11 3.6% 797 3.3% 1 0.4% 11 0.2% Bernhard Kettemann & Georg Marko 82 Mental 4 40 20.3% 6,244 26.2% 18 6.8% 259 4.0% Cognition 6 3.0% 1,166 4.9% 8 3.0% 94 1.4% Emotion 28 14.2% 4,350 18.3% 5 1.9% 93 1.4% Perception 3 1.5% 556 2.3% 3 1.1% 48 0.7% Science — — — — 6 2.2% 108 1.6% Geography/ politics 1 0.5% 92 0.4% 13 4.9% 257 4.0% Chemical — — — — 2 0.8% 47 0.7% Ecology — — — — 17 6.4% 680 10.5% Economy 3 1.5% 176 0.7% 50 18.9% 1,466 22.6% Computer 3 1.5% 303 1.3% — — — — Physical contact 4 2.0% 342 1.4% — — — — TOTALS 5 197 23,823 265 6,483 Table 3: Relative sizes of semantic domains with respect to lexical variation and token frequencies in the Emo corpus and the LOHAS corpus. The results reveal a lot about the nature of the two discourses represented in the corpora. On the one hand, we see domains that stress the personal, individual and subjective sphere of people on top in the Emo corpus, especially emotions, but also communication, the body, style and social relations. On the other hand, the LOHAS corpus has dimensions of general, collective and global relevance on top, with a strong lead by economy, but also ecology, geography/ politics (including references to the world, to countries, etc.) and quantity. This dichotomy between individual and global seems to corroborate the assumption that Emos and LOHAS supporters are alternative in a very different sense. Focusing on their own destiny, Emos seem to favour personal withdrawal as a form of being alternative. It might be objected that this personalization is less a result of the lifestyle standing behind the texts than a consequence of the inclusion of lots of personalized genres such as blogs. This is of course true. But we also have to consider that Emo discourse is strongly tied to such personalized genres, which means it does not happen in other forms, so that we cannot disentangle the two aspects. LOHAS, on the other hand, promotes a notion of alternative that focuses on the environment and on our wellbeing as general phenomena. Given the prominence of the economy as a semantic domain, this seems to be the main path to achieve such a reformist agenda. The selection of the texts, which partly verge on promotional discourse, plays a role in this. But there is no denying that the strong emphasis on the economic 4 The superordinate category also includes words that could not be clearly assigned to any of the subordinate classes, the numbers therefore do not represent the sum of the numbers of the individual classes. This also applies to the social class. 5 Even though the top 250 words from the wordlist were considered, the totals in the type column are not 250. The reason for this is that in the Emo corpus, there were a lot of lemmata, i.e. e.g. Emo, Emos and Emo’s count as one type. In the LOHAS corpus, on the other hand, which is much smaller in size, a large number of different words share the 250 th position, so we had to include all of these. The Language of Alternative Lifestyles 83 side also points in the direction of a capitalist conception of the world. And even though capitalism is definitely in the mix of the mainstream culture that alternative lifestyles seek to undermine, LOHAS’ Otherness may be superficial after all, focusing on the economic side of the consumption of the lifestyle signifiers. We do not deny that LOHAS efforts may have effects, but the question remains whether the conceptual basis does not rely too much on the mainstream to really make a long-term difference. The only category that ranks highly in both corpora is evaluation. Considering that alternative lifestyles - perhaps even more so than mainstream lifestyles - are concerned with ethical attitudes associated with particular signifiers, the salience of evaluative terms should not come as a surprise. Is there anything in the evaluative lexemes themselves that would point to a difference between the two routes of Emos and LOHAS? First of all, it has to be mentioned that evaluation may play an even larger role in the LOHAS discourse. The data shown in the table just relate categories to each other. But it does not show, for instance, that the LOHAS corpus actually features more different evaluative terms than the Emo corpus despite being smaller. When we look at the set of terms in the LOHAS corpus, we see that practically all of them denote positive evaluation. authentic; clean; efficiency; fair; fitness; free; good/ well; great; green; important; integrative; key; opportunity; primary; quality; significant; successful; top There are also positive evaluations in the Emo corpus. cool; deep; good; hot; matter; ok; pretty; right; top; true But, as was to be expected, there are also many negative ones (though they do not occur more often, perhaps slightly surprisingly). bad; hard; hell; loser; problem; shit; stupid; wrong We probably expect alternative lifestyle discourses to compare the mainstream to one’s own lifestyle on aesthetic and especially on ethic grounds. This implies that there should be both positive and negative evaluative elements in the text. This is indeed the case with the Emo discourse, but not with the LOHAS discourse. Could this indicate that LOHAS wants to praise itself, but not necessarily in comparison to the mainstream? This could, of course, be interpreted as meaning that LOHAS, as a matter of fact, already contains aspect of a mainstream lifestyle. It has to be mentioned in this context, though, that the promotional slant of some of the articles included in the corpus has contributed to this foregrounding of the positive. Bernhard Kettemann & Georg Marko 84 5.2 Keywords WordSmith also allows the production of comparisons between corpora, showing which words occur significantly more often in one corpus than in another. These words are defined as keywords. In the following table, you see comparisons between the Emo and the LOHAS corpora. The left column contains the words that most clearly distinguish the discourse of Emos from that of LOHAS, the right column contains the words that most clearly distinguish the discourse of LOHAS from that of Emos. 6 Emo vs. LOHAS LOHAS vs. Emo I LOHAS my the you of me organic Emo products am/ I’m green so consumers hair and love energy never business like companies by in just growth don’t environmental her consumer joined yoga back market she percent it sustainable think spa post health location company subject technology pain their really as music industry know medicine hate association 6 Most of the function words appearing in the lists point to differences related to style and genre rather than to the thematic perspectives proposed, e.g. personal and possessive pronouns, especially first and second person pronouns, indicate a more personal style in one of the corpora, while the definite article indicate a less personal style employing noun phrases rather than references to the speaker or the addressee. The Language of Alternative Lifestyles 85 broken says get clean mood research heart sustainability Table 4: Keyword comparison between the Emo and LOHAS corpora (left) and the LOHAS and Emo corpora (right). We have also compared the two corpora to a large general corpus of English. For this purpose, we have chosen to use the British National Corpus (= BNC), a 100-million-word corpus of British English, compiled in the first half of the 1990s and consisting of written (90%) and spoken (10%) material produced, roughly speaking, in the 1980s and 1990s (cf. Aston/ Burnard 1998). This is supposed to allow more general insights into which lexical items are key to the two discourses under scrutiny. Emo vs. BNC LOHAS vs. BNC Emo LOHAS I/ me/ my organic am/ Im consumers hair yoga you/ your green posted products posts spa love sustainable/ sustainability joined ecotourism post Fairmont location energy lol percent GBP 7 environmental/ ly hate antioxidant by growth pain medicine heart mega-banks mood Weil hearted companies broken clean like marketplace just NMI 8 Table 5: Keyword comparisons between the Emo corpus and the BNC and the LOHAS corpus and the BNC. All in all, the keyword comparisons support the interpretations from above. We again see words pointing to emotional individualization in the 7 giving prices of fashion items 8 = National Marketing Institute Bernhard Kettemann & Georg Marko 86 Emo corpus and political and economic collectivization in the LOHAS corpus. 5.3 Concordances and collocations Lists of words allow insights into which aspects of meaning are salient in a discourse. But being limited to isolated items of text, they can only tell part of the story and must be complemented with analyses that also take into account the combinations that words enter into. By producing concordances, which display a word (or a string of words) as it occurs in the text, i.e. in its immediate verbal context, we can examine such combinations. With the help of WordSmith’s Collocate function, we can additionally produce lists of the most common words occurring within a defined distance to the search word, thus getting immediate access to frequent collocations. 5.3.1 Collocations in the Emo corpus: Verbs following first person I As has been discussed above, the conception of the world on which the lifestyles of Emos is based shows a strong emphasis on the individual person. It therefore is interesting to see how this individual is constructed in the Emo discourse. One way of looking into this question is to examine the verbs collocating with the first person singular pronoun I as their subject. This will reveal in which events Emo individuals - we assume that the first person pronoun is the major linguistic element referring to Emos in the texts - participate. And what Emos say they do will be pivotal in their construction of their own identities. We limit the search to the first word directly following the first person singular pronoun. This means missing out on verbs of more complex verb groups (including negations, modals, semi modals, etc.), but the data would be very difficult to process otherwise. And we still think that the majority of verbs directly follow the pronominal subject. The resulting list of collocates is then categorized semantically. The categories we are using are the following. ‣ Existence & change: processes of being and becoming; e.g. exist, become. ‣ Possession: processes of having, getting or losing; e.g. own, obtain, lose. ‣ Movement & static position: processes of changing - e.g. walk, come - or keeping one’s position - e.g. lie, sit. ‣ Phases: processes of starting, finishing or continuing; e.g. stop, begin, keep on. The Language of Alternative Lifestyles 87 ‣ Physiology: active and reactive processes of the body; e.g. eat, sleep. ‣ Social processes: e.g. date, meet. ‣ Communication: processes of communication and interaction; say, promise, convince. ‣ Cognition: processes of thinking; e.g. know, remember. ‣ Perception: processes of sensual perception; e.g. see, hear, look at. ‣ Emotion: processes of affection, volition and evaluation; e.g. fear, enjoy, need. ‣ Style: processes concerned with dressing, make-up and hairstyle; e.g. wear, dye. ‣ Physical contact: processes of physical contact with people or objects; e.g. touch, cut, kiss. There are also verbs that could not easily be assigned to any of the categories. The table below contains all the verbs found (with the exception of the unclassified ones) in their respective categories. The numbers between brackets give absolute frequencies (higher than one). Existence & change: die (18); live (11); get (‘become’) (9); become (4); exist (3); change (2); go Emo (2); grow (2); disappear; turn (‘become’), pass away Possession: have (139); get (‘receive’) (42); give (20); take (12); lose (11); keep (4); belong; buy; gain; steal Movement & static position: go (18); walk (10); fall (9); come (7); move (4); run (3); crawl (2); get (somewhere) (2); skate (2); approach; bow; creep; glide; hop on; march; slide lie (‘horizontal bodily position’) (17); sit (10); lean (4); stand (3); kneel Phases: start (18); stop (9); end (3); keep (on) doing (3); launch (3); begin (2); give up (2); stay (2); finish; go on; keep up; resign Physiology: cry (42); bleed (6); wake (up) (5); fall asleep (4); sleep (4); laugh (3); awake (2); eat (2); collapse; drain; draw a breath; drink; drown; faint; starve; swallow; take a breath; take medication; weep Social processes: meet (8); help (2); break up; celebrate; date; go out; join Communication: say (48); write (37); tell (22); ask (16); scream (9); swear (6); bet (5); call (5); agree (4); pray (4); promise (4); beg (2); blame (2); explain (2); lie (‘not to tell the truth’) (2); plead (2); recommend (2); talk (2); admit; answer; apologize; convince; defend; disagree; mutter; preach; read; refuse; scribble; stutter; text; voice Cognition: think (127); know (122); guess (30); remember (19); mean (13); wonder (10); realise (9); believe (6); get (‘understand’) (6); dream (5); pick (4); plan (4); decide (3); doubt (3); learn (3); look back (metaphorically) (3); suppose (3); choose (2); expect (2); figure (2); reminisce (2); understand (2); con- Bernhard Kettemann & Georg Marko 88 fuse; consider; forget; recognize Emotion: want (109); feel (105); love (102); need (92); hate (71); like (54); wish (44); hope (29); care (14); miss (10); bottle up (5); dig (5); fall in love (4); fear (4); break (down) (3); fall for (3); hurt (3); adore (2); long for (2); bother; cherish; comfort; crave; dare; deal (‘cope’); dislike; dread; heart; look forward; pity; prefer; revel; suffer; take (‘bear’); take it to the heart; trust Perception: see (75); look (‘gaze’) (19); hear (17); watch (5); stare (4); listen (3); taste (3); gaze (2); glare; notice; peer Style: wear (26); look (‘appear visually’) (9); dress (8); dye (4); pierce (2); put on (2); braid; clip; color; put hair in a ponytail; sport Physical contact: cut (27); hold (14); tear (7); push (4); smash (4); stab (4); grab (3); press (3); rip (3); turn (3); beat (2); break sth. (2); give a kiss (2); hit (2); kick (2); kiss (2); pull sth. out (2); clench; clutch; embrace; grasp; grip; hug; press sb. close; pull; scratch; slash; slice; snap; snatch; squeeze; strike; touch; unwind; wipe; wrap Table 6: Major semantic classes of verbs following first person singular I in the Emo corpus. The percentages in the table below represent the sizes of the categories relative to each other and have been calculated in the same way as in 5.1 (Tables 3 and 4) above. All verbs have been included for the calculations, even if they could not be assigned to any of the classes mentioned. Types Tokens Existence & change 11 3.4% 54 2.3% Possession 10 3.1% 232 9.9% Movement & static position 21 6.5% 99 4.2% Phases 12 3.7% 46 2.0% Physiology 19 5.9% 79 3.4% Social processes 7 2.2% 15 0.6% Communicative processes 32 9.9% 188 8.0% Cognition 26 8.0% 384 16.4% Perception 12 3.7% 159 6.8% Emotion 34 10.5% 676 28.9% Style 11 3.4% 56 2.4% Physical contact 36 11.1% 105 4.5% TOTALS 324 2,339 Table 7: Absolute frequencies and relative sizes of the semantic categories of the verbs following first person I in the Emo corpus. The data in this table supports the conclusion from above about the prevalent importance of emotions for Emos - nomen sometimes still est omen. This category could probably be even further ahead if we consider that some processes categorized as related to the body, i.e. as physiologi- The Language of Alternative Lifestyles 89 cal, such as, for instance, cry or laugh (the category needs to be stretched to accommodate them, we admit), could be included as well. Certain processes, even though showing high frequencies, do not seem to be peculiar to the Emo discourse, e.g. cognitive processes (after all, this includes any instance of I think) or possessive processes (including any occurrence of I have as a lexical verb). The fact that communicative events feature relatively prominently, especially with respect to lexical variation, points to the dilemma of the Emo identity - lonely but reaching out for communicative contact. One category that is ranked surprisingly high in the type column is physical contact. Obviously, there is a great variety of terms that describe how Emos touch - however forcefully - themselves, other people or things. A look at the verbs shows overwording of those words that denote a strong, even violent, act by the first person. cut; tear; push; smash; stab; grab; press; rip; beat; break; hit; kick; clench; clutch; grasp; grip; scratch; slash; slice; snap; snatch; squeeze; strike Verbs such as embrace, kiss or hug are also included in the list, but are comparatively rare. This suggests that there is some moment of violence in what Emos do (according to their own descriptions). This could of course point to a more confrontational dimension of their Otherness. A closer look at the data, e.g. in a concordance, showing all the verbs mentioned in their immediate verbal context, however, reveals that many of the verbs occur in the description of auto-aggression and aggression against objects, but rarely directed against other people. instead I thought I would just cut myself till I felt better... it like REAL DEPRESSED and well I cut myself with a knife...I cam dep ent cut since then. Now i just scar myself. But lately I have star Post subject: well I don't cut myself but I don't judge ppl. I straight, just think shit? ! ? i tore my fucking heart open just for t manage to control myself. I cut myself open again, tears fallin wall, I swear I’m not insane beat myself up ‘till I fall, I hard ed, I feel like such a prat I smash my head against the wall, I s deep I feel like such a fool I slash at my flesh That I broke, tha I hate. I cried all day, now I cut my wrist. To end the pain the p 4 months straight... i mostly slashed my palms cuz i figured the Concordance 1: Violent verbs in auto-aggression of first person. These aspects thus add to the impression that Emos try to define their difference to the mainstream in solipsistic terms, focusing on themselves rather than being concerned with changing their mainstream environment. Bernhard Kettemann & Georg Marko 90 5.3.2 Collocations in the LOHAS corpus: consumer and product One of the most frequent content words in the LOHAS corpus is the noun consumer. Being interested in the collocations of the word, we have produced a list based on Mutual Information (= MI). This is a statistical measure defining the degree of association between two elements. It is not simply based on the frequency at which two words occur together (after all, nouns tend to occur with the definite article more often than with any other word and still we do not speak of a strong connection there), but it also includes the frequencies at which two words occur without each other. The more frequent the co-occurrences and the fewer the solitary occurrences, the higher the Mutual Information (cf. Oakes 1998: 63-65, Mukherjee 2009: 89f.). Here are the words that show a high MI score and can therefore be argued to be tightly associated with consumer in the LOHAS corpus. 9 consumer attitude consumer awareness consumer data consumer demand consumer education consumer expectation consumer issue consumer market consumer movement consumer product consumer report consumer research consumer sales consumer segment/ segmentation consumer target consumer trend Table 8: Combinations starting with consumer with a high Mutual Information. All of the collocates are nouns, which means that all combinations are as a matter of fact compounds. What could be the effect of this? Compounds may mitigate the personalization of consumers. Even though consumer on its own denotes a person, being part of a compound the noun is reduced to a modifier in words denoting non-personal concepts. This effect is enhanced by the fact that all of the compounds are expressions from marketing research, where personal psychology is transformed into ab- 9 It has to be mentioned, however, that the Mutual Information score is not sensitive to the frequencies of collocating words. This means that two words may have a high MI score even though one of them occurs just once in the corpus (cf. Mukherjee 2009: 104). The Language of Alternative Lifestyles 91 stract concepts, e.g. awareness, attitude, expectation, and behaviour is represented only as part of scientific practices, e.g. research, report, trend, etc. This adds to the strong emphasis on economic aspects and would thus be in line with the interpretations offered in the discussion of the most common semantic domains of content words. Let us look at the collocates of a second lexeme that appears frequently in the LOHAS corpus, viz. product, or rather the plural form products, which occurs significantly more often. This time, however, we will look for collocates preceding the search word. farm products (personal) care products building products green products organic products its products new products consumer products our products Table 9: Combinations ending in products with a high Mutual Information. The results underline the inconsistency of the LOHAS worldview already discussed above. On the one hand, most of the expressions denote goods produced according to ethically sound - ecologically speaking - principles (in this context, it must be mentioned that even words such as consumer products and building products are used exclusively with modifiers such as green or ecologically friendly). On the other hand, however, consumer product, new product and also the possessive pronouns its and our, which relate products to the companies making and selling them, emphasize the marketing dimension, marrying an ecological perspective with a capitalist slant. Conclusion This article set out to demonstrate that examining large samples of texts by Emos and proponents of LOHAS (= Lifestyle of Health and Sustainability) with a Critical Discourse Analytical approach and corpus linguistic methods could shed light onto the ways these two groups, which can both be defined as subscribing to alternative lifestyles, construct their Otherness, i.e. their deviance from a postulated mainstream, in their discourses. Although we confined ourselves to the analyses of word lists, keywords and verbs following first person singular I (in the Emo corpus) and compounds with consumer and product (in the LOHAS corpus), which Bernhard Kettemann & Georg Marko 92 means, as indicated, that the study can only be considered a preliminary pilot project, even this tentative look at results shows two very different approaches to alternative lifestyles. Emos seem to exclusively focus on themselves and their - mostly negative - emotions. They, however, are less concerned with a withdrawal rather than with a rejection of mainstream culture. LOHAS supporters, on the other hand, are reformist in orientation, seeking to improve the well-being of humans in balance with the well-being of non-human nature. They, however, seem to seek to achieve this goal partly through economic means. Thus agreeing with capitalist principles of mainstream culture could be argued to undermine the status of LOHAS as genuinely alternative rather than as an ‘ecologized’ version of conventional problem-solving strategies. References Archer, Dawn/ Andrew Wilson/ Paul Rayson (2002). “Introduction to the USAS Category System.” http: / / ucrel.lancs.ac.uk/ usas/ usas%20guide.pdf (23 Jan 2012). Aston, Guy/ Lou Burnard (1998). The BNC Handbook. Exploring the British National Corpus with SARA. Edinburgh: Edinburgh UP. Baker, Paul (2006). Using Corpora in Discourse Analysis. London: Continuum. Baker, Paul/ Andrew Hardie/ Tony McEnery (2006). A Glossary of Corpus Linguistics. Edinburgh. Edinburgh University Press. Chaney, David (1996). Lifestyles. London & New York: Routledge. Fairclough, Norman (1989). Language and Power. London & New York: Longman. Fairclough, Norman (1992). Discourse and Social Change. Cambridge: Polity. Fiske, John (1994). Understanding Popular Culture. London: Routledge. Goatly, Andrew (2000). Critical Reading and Writing. London & New York: Routledge. Kelley, Trevor/ Simon Leslie (2007). Everybody Hurts - An Essential Guide to Emo Culture. New York: Collins Harper Publishers. Kettemann, Bernhard (2011). “Tracing the Emo side of life. Using a corpus of an alterative youth culture discourse to teach Cultural Studies.” In: Ana Frankenberg-Garcia/ Lynne Flowerdrew/ Guy Aston (eds.). New Trends in Corpora and Language Learning. London & New York: Continuum. 44-61. Kirig, Anja/ Eike Wenzel (2009). Lohas: Bewusst grün - alles über die neuen Lebenswelten. München: Redline. LOHAS online (n.d.). http: / / www.lohas.com (17 Dec 2011). Marko, Georg (2008). Penetrating Language. A Critical Discourse Analysis of Pornography. Tübingen: Gunter Narr. Marko, Georg (2010). “Heart disease and cancer, diet and exercise, vitamins and minerals. The Construction of Lifestyle Risks in Popular Health Discourse.” In: CADAAD Journal (Critical Approaches to Discourse Analysis Across Disciplines) 4: 2 (Special Edition. Risk as Discourse. Ed. By Jens O. Zinn). 147-170. [online] http: / / www.cadaad.net/ 2010_volume_4_issue_2/ 65-53 (17 Dec 2011). Mukherjee, Joybrato (2009). Anglistische Korpuslinguistik. Eine Einführung. Berlin: Erich Schmidt Verlag. The Language of Alternative Lifestyles 93 Oakes, Michael P. (1998). Statistics for Corpus Linguistics. Edinburgh: Edinburgh UP. Tribble, Chris (1997). “Improvising corpora for ELT: quick-and-dirty ways of developing corpora for language teaching.” In: Barbara Lewandowska- Tomaszczyk/ Patrick James Melia (eds.). PACL 97: Practical Applications in Language Corpora. Łodz: Łodz University Press, 106-117. Software Scott, Mike (2008f.). WordSmith Tools 5.0. Bernhard Kettemann & Georg Marko Institut für Anglistik Karl-Franzens-Universität Graz