talen.en.tekens.book Page 1 Tuesday, November 10, 2015 11:55 AM
Over talen en tekens Bijdragen van jonge Gentse linguĂŻsten
talen.en.tekens.book Page 3 Tuesday, November 10, 2015 11:55 AM
Over talen en tekens Bijdragen van jonge Gentse linguĂŻsten
JOOST BUYSSCHAERT (red.)
talen.en.tekens.book Page 4 Tuesday, November 10, 2015 11:55 AM
Dit is een coproductie van Studia Germanica Gandensia (Libri) en Spieghel Historiael, het jaarboek van de Bond van Gentse Germanisten
© Academia Press P. Van Duyseplein 8 9000 Gent Tel. 09/233 80 88 Info@academiapress.be www.academiapress.be Uitgeverij Academia Press maakt deel uit van Lannoo Uitgeverij, de boeken- en multimediadivisie van Uitgeverij Lannoo nv.
Joost Buysschaert (red.) Over talen en tekens – Bijdragen van jonge Gentse linguïsten Gent, Academia Press, 2015, 94 pp.
Opmaak: punctilio.be Cover: Studio Eyal & Myrthe
ISBN: 978 90 382 2569 2 D/2015/4804/168 NUR1 616
Niets uit deze uitgave mag worden verveelvoudigd en/of vermenigvuldigd door middel van druk, fotokopie, microfilm of op welke andere wijze dan ook, zonder voorafgaande schriftelijke toestemming van de uitgever.
talen.en.tekens.book Page 1 Tuesday, November 10, 2015 11:55 AM
Inhoudsopgave
INLEIDING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . JOOST BUYSSCHAERT
3
A LINGUIST’S DAY ON A PLATE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Exploring the Rhetorical Resultative in the Dietary Industry JESSE BERWOUTS
5
DEUTSCHE UND NIEDERLÄNDISCHE KONDITIONALSÄTZE MIT SOLLTE(N) UND MOEST(EN) / MOCHT(EN). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Eine sprachvergleichende Korpusstudie SARA DELVA
29
DE SEMANTISCHE EVOLUTIE VAN DE ZESTIENDE-EEUWSE AAN-CONSTRUCTIE Een semasiologisch en onomasiologisch onderzoek DARIO RENS
57
JESUS IN SCANDINAVIAN RUNIC INSCRIPTIONS . . . . . . . . . . . . . . . . . . . . . DOESJKA TILKIN
77
PERSONALIA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
95
1
talen.en.tekens.book Page 3 Tuesday, November 10, 2015 11:55 AM
Inleiding JOOST BUYSSCHAERT
Het Spieghel Historiael Jaarboek van de Bond van Gentse Germanisten presenteerde jarenlang themanummers. Sinds 2012 is gekozen voor een andere formule: geselecteerde recent afgestudeerden van de Bachelor of Master worden uitgenodigd om een artikel te schrijven op basis van hun scriptie. De kwaliteit van deze bijdragen is telkens verbazend sterk. Dit is ook nu het geval. Na de literaire scripties waren dit jaar de linguïstische aan zet. Jesse Berwouts onderzocht de sloganeske constructie van het type “eat yourself thin”. Resultatieve constructies bestaan al langer, maar hier gaat het om “rhetorical resultatives”. Wat er eigen aan is, onderzoekt Jesse zowel vanuit het oogpunt van de pragmatiek als dat van de constructiegrammatica. Taaladviseurs hebben (blijkbaar vruchteloos) jaren gestreden tegen het Belgisch-Nederlandse “als hij dat moest zien”, maar “moest” kun je in deze constructie niet altijd vervangen door “mocht”. Sara Delva gebruikt fijnmazige technieken om het gebruik te vergelijken van “moest”, “mocht” en het Duitse “sollte” in voorwaardelijke zinnen. Door “deflexie” is het Nederlands gaandeweg zijn datief gaan vervangen door een constructie met het voorzetsel “aan”. Dario Rens situeert de (snelle) overgang in de zestiende eeuw met overtuigende corpusvoorbeelden en laat verstaan dat de evolutie eerst is gestart bij werkwoorden waar “aan” letterlijker kan worden opgevat en daarna is uitgebreid.
3
talen.en.tekens.book Page 4 Tuesday, November 10, 2015 11:55 AM
OVER TALEN EN TEKENS
De bijdrage van Doesjka Tilkin laat ons proeven van de runologie. (Het zou wijlen Professor Derolez, bekend Gents runoloog, een genoegen hebben gedaan dat jonge vorsers zich weer met runen inlaten.) Zij focust op de spelling van de naam Jezus, waarvan de beginletter vaak met de rune voor g wordt gespeld in de plaats van met het teken voor i. Zij probeert te ontwarren waar en wanneer die alternatieve spelling wel en niet voorkomt. “Taal en teken” is een bekende collocatie, maar in dit Jaarboek gaat het over meer dan één taal en meer dan één teken. Daarom kreeg de bundel de naam “Over talen en tekens” mee. De bijdragen kwamen onder tijdsdruk tot stand en hier past het de jonge onderzoekers te danken voor hun inzet om toch de krappe timing te respecteren. Ook de promotoren van de scripties waaruit de artikelen zijn afgeleid, verdienen dank voor hun aandeel in het selecteren en begeleiden van de auteurs. Het gaat om de professoren Miriam Taverniers, Anne Breitbarth, Timothy Colleman en Jóhanna Barðdal.
4
talen.en.tekens.book Page 5 Tuesday, November 10, 2015 11:55 AM
A Linguist’s Day on a Plate Exploring the Rhetorical Resultative in the Dietary Industry JESSE BERWOUTS
1.
Introduction
Have you ever eaten yourself thin? Neither have I. In fact, I eat like a pig and have the Body Mass Index of one. And yet, magazines and other media occasionally compel us to ‘eat ourselves lean’ and ‘sleep ourselves slim’. The average speaker intuitively detects ‘some kind of anomaly’ in such sentences: how can eating, an activity that involves stuffing oneself with calories rather than burning them, make someone thin? In this paper, I will argue that this counterintuitive nature of these sentences is used for their rhetorical effect related to their particular context. I have therefore labelled this category rhetorical resultative. Further sections will provide ample examples, but I will give a few introductory examples for the sake of demarcation: (1) a. Good smells are making you fat! STINK yourself thin b. Hi, I’m Troy McClure. You may remember me from such self-help videos as Smoke Yourself Thin and Get Confident, Stupid. c. Swim yourself slim We immediately notice a difference between (1a-b) and (1c): it would not be that hard to imagine a situation in which you become slim by swimming. Becoming thin by stinking, on the other hand, seems much harder to accept. We can distinguish a few shared features of (1a-c). I will only come to a ‘strict’ definition after having dealt with pre-existing theories and further
5
talen.en.tekens.book Page 6 Tuesday, November 10, 2015 11:55 AM
OVER TALEN EN TEKENS
examples from a previous corpus study. For now, these characteristics of (1a-c) will suffice: (i). their mood is the imperative, (ii). they appear in a context in which some sort of methodology is offered to achieve a certain result, (iii). they appear in short, eye-catching contexts such as headlines of media articles, (iv). a counterintuitive relationship between verb and result is not compulsory, but it is not impossible or even uncommon either. When we arrive at the point of a thorough definition, I will recapitulate these four characteristics, among other shared features of the examples, in contrast with the ‘mainstream’ resultative construction (which will be discussed simultaneously). For this research, ‘mainstream resultatives’ can be considered sentences imposing a result on an entity by means of an action, that are structurally and semantically similar to (1a-c)1. For example, the border collie licked its paw clean: by licking its paw, the border collie makes it clean. This paper aims to systematically explore this ‘rhetorical resultative’ (RR) and define it by means of strongly correlating features, found in sentences from a Google corpus. I will argue that the RR is both more and less productive than the ‘general’ resultative construction (RC), since it does allow some deviations from the RC’s restrictions, but is bound to specific pragmatic contexts. I will start by presenting the corpus research (limiting myself to the dietary results ‘X yourself thin/ slim skinny’) and discuss anomalies found in their lexical distributions. These anomalies will be contrasted with some restrictions of the RC in section 3, which will lead us to a definition of the RR in opposition to the RC in Construction Grammar (CxG) analyses. Section 4 will elaborate on the specific contexts in which the RR appears, by discussing its pragmatics from a text linguistics and marketing perspective. This article is based on a more extensive corpus study than is presented here. A more elaborate paper is available on academia.edu (Berwouts, 2015).
2.
Corpus research: the rhetorical resultative put to practice
2.1. Introduction and methodology As mentioned in the introduction, resultatives impose a result on an entity by means of an action. This section will quantify and discuss the lexical distribu1. The resultative construction is taxonomically broader than the structure in the examples; for instance, some are intransitive and indicate a change of state in the subject rather than the object (e.g. the pond froze solid). However, those other type are of no concern to the rhetorical resultative.
6
talen.en.tekens.book Page 7 Tuesday, November 10, 2015 11:55 AM
A LINGUIST’S DAY ON A PLATE
tion of verbs in three resultative templates ‘X yourself Z’, Z being the results thin, slim and skinny, and X being the different verbs occurring in the templates. The tables of the verbs, along with their features and frequencies, will show some systematic anomalies corresponding to the features distinguished in my initial examples (e.g. the imperative, headline-ish nature and a counterintuitive relationship between the verb and the result). Obviously, the methodological and, more specifically, dietary context is already implied by the three adjectives I will limit myself to. These lexical distributions are the result of corpus research on the search engine Google, by entering ‘X yourself thin/slim/skinny’ and quantifying the verbs along with some relevant features. These features are [-imp] and [-pol], respectively expressing any other mood than the imperative and negative polarity (i.e. negation, cf. I didn’t eat myself thin). I assume by default that a prototypical instance of the RR is in the imperative mood and has positive polarity, so only the negative occurrences per verb will be quantified. These tables will be shown and discussed in 2.2, while 2.3. will distill a general pattern found in these templates. This general pattern will be compared with CxG theories regarding resultatives in section 3, to come to a stricter definition of the RR. Note that I will not label these corpus sentences as instances of the RR yet. In the introduction of this paper, I suggested the imperative mood as one of the criteria in my proto-definition of the RR. However, I will treat the nonimperative corpus data equally at this point. I consider it of utmost importance to include the frequency of non-imperatives in my quantified data. Ignoring them would lead to an unwarranted restriction of the delineation of the RR’s domain, since some adjectives or verbs may co-occur systematically with a non-imperative mood. This hypothetical simultaneity, along with other co-variations, may bring us to a more fine-grained definition of the RR. I will use the corpus data in this section to gradually build towards a definition of the RR. Distinguishing a general pattern in 2.3. will be a logical next step towards a definition. 2.2. The data: X yourself thin, slim and skinny Using the legend from 2.1, quantitative results for the three templates can now be presented and compared. Because of their strong similarity, I will discuss the templates simultaneously, after the presentation of the results. The discussion will be subdivided into types of verbs, productivity/creativity and some quantitative features.
7
talen.en.tekens.book Page 8 Tuesday, November 10, 2015 11:55 AM
OVER TALEN EN TEKENS
2.2.1. X yourself thin Table 1: X yourself thin Verb Think Cook Eat Love Walk Visualize Coach Dress Sleep Dance Psych Smoke Relax Hypnotize Stretch Tax Write Snack Scare Dream Drink Spin Nudge Program Tap Wrap Organize Shag Stink Pyt pole Color Poop Picture Grin Pray Exercise Brainwash Bitch Brag
8
Frequency /154 36 20 13 7 5 4 4 3 3 2 2 2 2 2 2 2 2 2 2 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
- Imp 10
- Pol
1
1
1
1
1
1
1
1
talen.en.tekens.book Page 9 Tuesday, November 10, 2015 11:55 AM
A LINGUIST’S DAY ON A PLATE
Table 1: X yourself thin (continued) Verb Tan Seks Heal Run Bore Wog Chew Hate Socialize Type Sad Shiver Keep Trick Tempt Click Text
Frequency /154 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
- Imp
- Pol
1
1
1
1
2.2.2. X yourself slim Table 2: X yourself slim Verb Eat Think Sleep Dress Love Snack Walk Wrap Imagine Dance Cook Swim Drink Stink Talk Skip Shop Get Juice
Frequency /263 36 30 19 11 10 10 9 7 7 7 7 6 6 5 5 4 4 4 3
-Imp 5 4 2 4 1 2 2
-Pol
1 1 1 1 2 2
9
talen.en.tekens.book Page 10 Tuesday, November 10, 2015 11:55 AM
OVER TALEN EN TEKENS
Table 2: X yourself slim (continued) Verb Keep Run Style Hula hoop Shake Adore Shag Hypnotize Stretch Laugh Make Hoop Cycle Dream Shiver See Snooze Sip S.A.S.S. Boogie Nurture Morph Zinc Believe Strut Jog Hate Exercise Text Nourish Bike Jump Snooze Row Google Slurp Sweat Bounce Taste Relax Tweat
10
Frequency /263 3 3 3 3 3 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
-Imp 1
-Pol
2
1
1
1
1 1
1
talen.en.tekens.book Page 11 Tuesday, November 10, 2015 11:55 AM
A LINGUIST’S DAY ON A PLATE
Table 2: X yourself slim (continued) Verb Pray Psyche Tape Soup Zumba Stuff Blog Trick Cheat
Frequency /263 1 1 1 1 1 1 1 1 1
-Imp 1
-Pol
1
2.2.3. X yourself skinny Table 3: X yourself skinny Verb Eat Sleep Laugh Drink Dress Think Wrap Hypnotize Snack Scare Spray Starve Organize Substitute Dance Sing Psych Style Color Shiver Breathe Shit Bore Love Train Poop Spray tan
Frequency /135 33 12 10 9 6 5 5 4 4 4 3 3 2 2 2 2 2 2 2 1 1 1 1 1 1 1 1
-Imp 5 2 2 1 3 2
-Pol
1 1 1
1
1
1
11
talen.en.tekens.book Page 12 Tuesday, November 10, 2015 11:55 AM
OVER TALEN EN TEKENS
Table 3: X yourself skinny (continued) Verb Skip Stripe Stir-fry Make Tense Tap Relax Swim Squeeze Tech Sip Shake Run Dream Grill
Frequency /135 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
-Imp
-Pol
1
2.2.4. Types of verbs The templates display some seemingly uncommon features; the distributions clearly show a majority of verbs having a counterintuitive relationship with the results. Sport-related verbs, which could intuitively be associated with thin, slim and skinny, are relatively rare (row, bike, run, swim, …). Some types of counterintuitive verbs seem to appear more frequently and diversely, e.g. nutrition (eat, cook, grill, juice, soup, …) and mindfulness (think, dream, hypnotize, love, …). We can distinguish five major lexical categories of verbs occurring with the templates: (i)
Verbs expressing behavior towards nutrition: eat, drink, cook, snack, taste, … (ii) Verbs of motion and sports: walk, run, dance, exercise, coach, shag, … (iii) Mental verbs that can be further subdivided into general verbs of cognition, mental well-being and emotion (think, love, relax) and verbs of controlled perception. These verbs of controlled perception express a voluntarily cogitated mental image of yourself as being thin, which is supposed to help you to become thin; e.g. visualize and picture. One instance of visualize yourself thin in the data occurs as the title of a website which explains the process as follows: ‘Visualization works for everything else. Why not visualize your perfect body – thin and fit and toned? As you visualize your perfect body at your 12
talen.en.tekens.book Page 13 Tuesday, November 10, 2015 11:55 AM
A LINGUIST’S DAY ON A PLATE
ideal weight, you are actually creating it. In accordance with the Law of Attraction, the process for attracting and creating your ideal body is the same as for attracting and creating anything you desire – focus on what you want until you bring it into being. As you visualize yourself thin, voila, you will be thin!2‘ (iv) Verbs involving other bodily processes, some being closely linked to mental states: sleep, poop, grin, heal, … (v) Verbs related to multimedia and social media: program, click, text, … There is a clear difference between (i-ii) and (iii-v) regarding intuitive relevance to thin, slim and skinny: verbs of nutrition and motion can intuitively be associated with the three results, either in a positive or negative way. Mental verbs, bodily processes and multimedia, on the other hand, are usually not associated with the results at all. The categories of nutrition and sports can be considered macro-categories of thin, slim and skinny: both categories are intuitively associated with the result, although eating is generally associated in a negative way. Therefore, the counterintuitive nature of eat yourself thin differs from blog yourself thin’s: eat is associated with a result contrary to thin (fat), while blog is not associated with the result at all. The results show that both types of counterintuitive relationships are more common than the positively associated macro-category of motion and sports. The tables also show many verbs falling outside of these five categories, and thus outside of the dietary macro-categories, co-occurring with the template: tax, dress, stink, color, … These verbs often apply new knowledge or contexts to construe a new reality: the article titled Scare Yourself Thin discusses scientific evidence of the link between horror movies and burning calories faster, while Tax Yourself Thin in 30 Days or Less suggests higher taxes on fast-food. The majority of verbs in these tables are verbs with a clear lexical meaning that usually do not appear with a resultative adjective. For instance: eat only takes a direct object, in most cases. However, the verbs make, get and keep are exceptions here: those three verbs are lexically void, and only serve to establish the resultative relationship. While these inherently resultative verbs are uncommon in the templates, they are relatively more common in ‘normal’ language and hence – because of their inherently resultative nature – much closer to resultative relations. The non-inherently resultative verbs in these tables establish a resultative relation via a means-relation (Goldberg, 1997: 386-387). For instance, eating is presented as a specific means of making 2. Source: http://thinkingthin.weebly.com/.
13
talen.en.tekens.book Page 14 Tuesday, November 10, 2015 11:55 AM
OVER TALEN EN TEKENS
yourself thin. It goes without saying that inherently resultative verbs cannot be in a counterintuitive relationship with the result, since their only meaning is establishing the result. This accounts for the general dominance of inherently resultative verbs in resultative construals (in the extended version of this paper, ‘X yourself rich’ had an inherently-resultative verb rate of 20%, and other templates were almost entirely inherently resultative). Thus, the systematic absence of inherently resultative verbs in these three templates is somewhat odd. A comparison with other quantifiable features in 2.2.6. will explore if this absence co-varies with other relevant features of the RR. 2.2.5. Productivity and creativity The long lists of verbs and diverse categories in these tables imply that these templates allow a lot of creativity; a language user can easily think of a verb that is not present in the list and analogously fuse it with one of the templates. For instance, telekinesis yourself thin would not sound much less acceptable than think or hypnotize; telekinesis does not belong to thin’s macro-category either, but it would be equally contextually justifiable (for instance, using your mind powers to scorch away your belly fat, or moving your limbs without neurological muscle contraction). If the empty slot in the template can be filled by a wide and diverse array of words (here: verbs), the template can be considered to be productive (for instance, in a CxG sense, cf. Goldberg, 1995). A common way of estimating a template’s productivity is by calculating the hapax ratio: the proportion of unique verbs (i.e. verbs that appear only once) to the total number of different verbs of a template. Thin’s hapax ratio is 62.5%, while the other two have hapax ratios centered around 50%. This is significantly high, indicating that the templates are quite productive. Telekinesis and some of the actual hapax verbs illustrate another facet of productivity and creativity: the templates allow neologisms. For instance, telekinesis would be a denominal verb, similarly to zinc yourself slim and pyt pole yourself thin. Another way of introducing neologisms is by means of portmanteaus: blending together two different verbs to create a new one. Examples are tweat yourself slim (tweet + eat) and wog yourself thin. Tweat illustrates the productivity of the diverse lexical categories to a further extent, combining the nutrition and multimedia categories. 2.2.6. Quantitative features In the previous subsections, I discussed some features that can easily be quantified. For instance, the proportion of inherently resultative verbs can easily 14
talen.en.tekens.book Page 15 Tuesday, November 10, 2015 11:55 AM
A LINGUIST’S DAY ON A PLATE
be acquired by counting the inherently resultative verbs in a template, and dividing this number by the total number of verbs. The same can be done for the proportion of imperatives, the proportion of counterintuitive verbs (although this remains subjective to a certain extent) and the hapax ratio. I have quantified these features for the three templates, and represented these data in percentages in the following models:
Figure 1: X yourself thin
Figure 2: X yourself slim
15
talen.en.tekens.book Page 16 Tuesday, November 10, 2015 11:55 AM
OVER TALEN EN TEKENS
Figure 3: X yourself skinny
As we can see, the four variables are similar for the three templates. The high hapax ratios indicate a high level of productivity. Contrary to ‘regular’ resultative corpus material, inherently resultative verbs are extremely rare. Counterintuitive relations between the verb and result are common, and so are imperatives. The stability of these four variables is a first step to distinguishing overarching features, which can be contrasted to the RC. Note that only the counterintuitive relationship breaches the ‘rules’ of the RC. The other variables only show co-varying tendencies, which may be attributed to the RR. These variables can be further compared in several ways. For instance, the few inherently resultative verbs present in the tables have a relatively high proportion of non-imperatives, which may indicate a correlation between both variables. The other templates in the full version of this paper suggest that this is indeed the case. However, proving a correlation between these four variables (i.e. systematically having higher and lower values together from template to template, indicating a strong positive correlation) would require long, complex and arduous multivariable analysis. Multivariable analysis his clearly lies beyond the scope of this paper’s purpose of illustrating the RR, but it is on the agenda for future extension of this research.
16