Plural mass nouns and the construal of individuation: cross ...

Plural mass nouns and the construal of individuation: cross ...

Plural mass nouns and the construal of individuation: cross-linguistic evidence from verbal and non-verbal behaviour in linguistic and non-linguistic contexts Athanasiadou, Ifigeneia Phd Candidate School of English Thessaloniki Cognitive Linguistics Reading Group Aristotle University and Dr. Athanasopoulos, Panos 1 Reader University of Reading, United Kingdom Introduction Testing for the linguistic relativity hypothesis (Whorf 1956), previous studies have found that speakers of different languages conceptualize objects and substances in different ways (Lucy 1992; Imai and Gentner 1997; Athanasopoulos 2006, 2008; Sera and Goodrich 2010). Studies conforming with the universal ontology position

have revealed that, even when a language does not grammatically encode the count/mass noun distinction, speakers, by default, categorise objects by shape and substances by material (Soja et al. 1991; Imai and Gentner 1997; Athanasopoulos 2006, 2008). 2 Count/mass noun distinction Conforming with Radden and Dirven (2007:64), objects can be discriminated from substances by relying on three complementary conceptual criteria: i) boundedness, ii) internal composition, and iii) countability. Objects and substances can also be distinguished on a number of grammatical attributes as well (Radden and Dirven 2007:66). For example, count nouns, unlike mass nouns, accept plural marking. In the study, the ontological distinction between objects and substances, as being encoded in the grammatical count/mass noun distinction, is examined in two languages, namely in Greek and in English, with

respect to the way this distinction is conceptualised and then expressed by speakers of the two corresponding 3 languages. Comparing Greek with English, the grammatical count/mass noun distinction is displayed in both languages (Tsoulas 2006, 2007; Athanasopoulos 2008; Athanasiadou 2009). A binary number system is applied to both languages, where count nouns obtain both a singular and a plural form and can directly be preceded by numerals. Regarding mass nouns, they behave similarly in Greek and in English, in that numerals cannot precede some instances of mass nouns (e.g., ?Arasa o zo zxares ?I bought two sugars). The relevant numeral needs to be followed by a kind of unitizer in order for the mass noun to be quantified (e.g. Arasa o zo kouti zxari I bought two boxes of sugar). 4

Mass noun pluralisation Mass nouns in Greek are regularly and systematically marked for plurality, unlike mass nouns in English. However, the existence of plural mass nouns in Greek is not a superficial linguistic convention, rather it contributes meaning. The relevant meaning is that of there is more amount of substance than what would normally be expected or needed (Tsoulas 2007:18): rixes nerrixes ner sto ptoma ?You spilt waters on the floor (Tsoulas 2007:4) Gmises to spti me xmata ?You filled the house with earths (Tsoulas 2007:4) These contexts do not involve substance-to-object shifts, named after Radden and Dirven (2007:71), since water/earth is neither measured in portions nor different varieties of water/earth are implied. 5 Linguistic Task: Greek Current study and English monolingual native adult speakers first complete a picture-naming task with familiar count and mass nouns from the domain of food and drinks.

Pairs of pictures are presented in one of three syntactic conditions: the informative-syntax (a glass/some water), the neutral-syntax (the glass/the water), and the no-label condition (no syntactic cues or labels). Potential pluralisation of count and mass nouns is examined in the two languages. Prediction 1: Pluralisation of count nouns will systematically be applied to both languages, irrespective of syntactic context. Prediction 2: Greek, when compared to English, speakers, will pluralise mass nouns more in all of the syntactic conditions. In Greek, thus, the grammatical count/mass noun distinction is hypothesized to be less strictly defined than in English. 6 Results

Proportion of plural marking for count and mass nouns by the two language groups across the syntactic contexts Plural count nouns Plural mass nouns 100 100 Informativ e syntax 80 40 Neutral syntax 40 20 No-label 80 60

0 The 60 20 0 Greek English Greek English results of the verbal task clearly show that both language groups significantly distinguish between count and mass nouns , confirming Tsoulas (2006, 2007), since they all gave a significantly greater amount of plural responses for count nouns than mass nouns across the syntactic conditions. 7 Results Proportion of plural marking for count and mass nouns by the two language groups across the syntactic contexts Plural count nouns 100 nouns 80

Informativ e syntax Neutral syntax No-label 60 40 20 0 Greek English Plural mass 100 80 60 40 20 0 Greek English

The two language groups pluralised count nouns to a similar extent across the syntactic contexts. However, Greek speakers pluralised mass nouns significantly more than English speakers in neutral 8 syntax as well as in a non-linguistic context. Count/mass noun differentiation Concerning count nouns, our prediction, that count nouns will be pluralised to a similar extent by Greek and English speakers across syntactic contexts, was hence confirmed. Regarding mass nouns ,Greek, when compared to English, speakers pluralised mass nouns to a greater extent in neutral syntax as well as in nolabel. Our second prediction was thus also confirmed, except when informative syntax was provided. Athanasopoulos (2008) is thus supported in claiming that the count/mass distinction in nouns is less strictly marked in Greek than in English.

9 The question hence should not be on whether mass nouns are linguistically marked for number or not within a language group, but on the degree and regularity with which mass nouns are pluralised across language groups and syntactic contexts. Regarding syntax, informative syntactic cues invite speakers to render the appropriate grammatical meaning. That is, count syntax enhances pluralisation of count nouns whereas mass syntax inhibits pluralisation of mass nouns. However, ambiguous syntax or non-existent syntactic cues/labels, allow for cross-linguistic grammatical differences to be revealed. 10 Seeking evidence for the linguistic relativity hypothesis, (Whorf 1956), which claims that language affects cognition, we cannot rely on verbal data

alone, but we also need independent measures of non-verbal cognition. We hence conducted a non-verbal cognitive task as well. 11 Cognitive task After participating in the linguistic task, monolingual native adult Greek and English speakers also complete a cognitive task, that is a triads matching task. They are presented with novel object and substance entities and are asked to choose between a shape or material alternate the one that best resembles the target entity. Each participant is provided with one of three syntactic contexts: an informative-syntax condition (a/some fep), a neutral-syntax

12 Figure 1. Example triad from the objects condition (a/the fep) Figure 2. Example triad from the substances condition (some/the fep) 13 It is typically assumed that, where there is a count/mass noun distinction in the grammar of a language, subjects will generalise to new instances by shape, when they see an object, and by material when they see a substance (Lucy & Gaskins 2001). Prediction 1: Both language groups will categorise objects similarly, that is by shape, since they similarly pluralised count nouns in the linguistic task. Prediction 2: However, since Greek, compared to English, speakers pluralised mass nouns more in the linguistic task, Greek speakers will categorize substances by shape to

a greater extent than English speakers. 14 Results Proportion of shape preferences for objects and substances by the two language groups across the syntactic contexts Object shape 100 100 80 80 Informativ e syntax 60 60 40 Neutral syntax

40 20 No-label 20 0 Greek English Substance shape 0 Greek English Both Greek and English speakers categorized objects significantly more by shape and substances significantly more by material in informative syntax.

Our first prediction, regarding a similar preference for object 15 shape by the two language groups, was hence confirmed. Results Proportion of shape preferences for objects and substances by the two language groups across the syntactic contexts Object shape Substance shape 100 100 80 80 Informativ e syntax Neutral syntax No-label

60 40 20 0 Greek English 60 40 20 0 Greek English However, a greater preference for substance shape by Greek, when compared to English, speakers was not observed. Our second prediction was hence not confirmed. Nevertheless, Athanasopoulos (2008) found that, in a non-linguistic context, Greek speakers selected substance shape more, when compared to English speakers. 16

Comparison of the two tasks The linguistic and the cognitive task were compared using multiple regression analysis, comparing the two language groups across the two tasks by collapsing the syntax conditions within each language group. The independent variables were plural responses for count nouns and plural responses for mass nouns and the dependent variable was shape scores for objects or shape scores for substances. 17 Greek speakers When the dependent variable was shape scores for substances, the overall regression was significant, F (2, 33) =5.85, MSE=1.50, p<.01 (R2=.26). The beta weight for mass nouns (=.51) obtained a significant value (p<.01), showing that the more Greek participants pluralized mass nouns in the linguistic task,

the more they categorized substances by shape in the cognitive task. Table 1.Count and mass noun plural production as predictors and substance shape as dependent variable across the two tasks for Greek speakers. B SE B Constant .33 3.00 Mass nouns .16 .05 .51* Count nouns -.01 .25

-.01 Note: *p<.01. 18 English speakers When the dependent variable was shape scores for substances, the overall regression was not significant, F (2, 33) =1.67, MSE=1.43, p>.05 (R2=.09), suggesting no causal relationship between count/mass noun plural production and preference for substance shape by English speakers. Table 2.Count and mass noun plural production as predictors and substance shape as dependent variable across the two tasks for English speakers. B SE B 1.40 .70 Mass nouns

.15 .10 .30 Count nouns -.05 .06 -.15 Constant 19 Conclusions for the two tasks In the linguistic task and in neutral syntax as well as in a non-linguistic context, Greek speakers pluralised mass nouns more than English speakers. Greek speakers differentiated hence between count and mass nouns to a smaller extent than English speakers. Support for the

linguistic relativity position is hence supplied. In the cognitive task and in informative syntax, the two language groups performed similarly by selecting shape for objects and material for substances. Support for the universal ontology position is, thus, provided. However, as the regression analyses revealed, the tendency to pluralize mass nouns in Greek predicts the tendency by Greek speakers to categorize substances based on shape. This clearly shows that the cognitive behaviour of speakers is influenced by the grammatical structure of their language as well, providing further support for the linguistic relativity hypothesis (Whorf 1956). 20 The findings of the linguistic task, namely that Greek, when compared to English, marks the count/mass noun distinction less strictly in its grammar, are hence supported by the results of the regression analysis, where mass noun pluralisation in Greek predicts the cognitive preference for substance shape by Greek

speakers. Informative syntax has been proved to coordinate language-specific with universal categorization patterns, since the two language groups, in both tasks, performed similarly, when informative syntactic cues were provided. 21 General conclusions Metonymic relationships source from our conceptualising system and lead to the employment of metonymic linguistic expressions (Kvecses 2010), such as tra ner three waters. In the study, Greek speakers systematically employed such metonymic expressions and this linguistic patterning infiltrated their conceptualisation of substances as if they were objects. Linguistic convention has also been argued to mark crosslinguistic category boundaries (Athanasopoulos 2008), like the convention of extensive mass noun pluralisation observed in Greek. The latter convention has hence

proved sufficient to fine-tune the non-verbal categorisation preferences of Greek speakers. 22 The findings of the present study converge with the view that the cognitive behaviour of speakers of a language is governed neither completely by ontological universalism nor completely by linguistic relativism, but by an inbetween state (Imai and Gentner 1997; Imai 2000; Imai and Mazuka 2003; Athanasopoulos 2008). The interrelation between innate knowledge and grammatical structure and of the latter two variables with real-world composition could lead to a better articulated knowledge of the human conceptualising system (Imai 2000; Imai and Mazuka 2003). 23 Research in crosslinguistic first language acquisition Do young children, acquiring Greek, also pluralise mass nouns? To what extent are their non-verbal categorisation

preferences influenced by universal ontological knowledge and to what extent by the specific linguistic conventions of the language they are acquiring? Are Greek and English children significantly differentiated in their construal and expression of individuation? 24 Future Research When English children overgeneralise the plural morpheme from count to mass nouns, as part of acquiring the plural, to what extent do Greek children also display this pattern, as part of acquiring the plural, and to what extent as part of hearing plural mass nouns, extensively employed in the input they receive? 25 References Athanasiadou, I. 2009. Count and Non-count Nouns in Greek and in English. Unpublished Master Thesis. United Kingdom: University of Essex.

Athanasopoulos, P. 2006. Effects of the grammatical representation of number on cognition in bilinguals. Bilingualism: Language and Cognition, 9, 89-96. Athanasopoulos, P. 2008. Universal and language-specific patterns of categorization. Glossa, 4, 51-72. Imai, M. 2000. Universal ontological knowledge and a bias toward language-specific categories in the construal of individuation. In S. Niemeier and R. Dirven (Eds.), Evidence for linguistic relativity (pp. 139160). Amsterdam/ Philadelphia: John Benjamins Publishing Company. Imai, M., and Gentner, D. 1997. A crosslinguistic study of early word meaning: universal ontology and linguistic influence. Cognition, 62, 169200. Imai, M., and Mazuka, R. 2003. Re-evaluating linguistic relativity: Language-specific categories and the role of universal ontological knowledge in the construal of individuation. In D. Gentner and S. GoldinMeadow (Eds.), Language in mind: Advances in the study of language and thought (pp. 429-464). Cambridge MA: MIT Press. 26

Kvecses, Z. 2010. Metaphor: A practical introduction. 2nd edition. New York: Oxford University Press. Lucy, J. A. 1992. Grammatical categories and cognition. A case study of the linguistic relativity hypothesis. Cambridge: Cambridge University Press. Lucy, J. A., and Gaskins, S. 2001. Grammatical categories and the development of classification preferences: A comparative approach. In M. Bowerman and S. C. Levinson (Eds.), Language acquisition and conceptual development (pp. 257- 283). Cambridge: Cambridge University Press. Radden, G. and Dirven, R. 2007. Cognitive English Grammar. Amsterdam/Philadelphia: John Benjamins Publishing Company. Sera, M. D. and Goodrich, W. 2010. Who thinks that a piece of furniture

refers to a broken couch? Count-mass constructions and individuation in English and Spanish. Cognitive Linguistics, 21, 419-442. Soja, N. N., Carey, S., and Spelke, E. 1991. Ontological categories guide young childrens inductions of word meaning: Object terms and substance terms. Cognition, 38, 179211. Tsoulas, G. 2006. Plurality of mass nouns and the grammar of number. Paper presented at the 29th GLOW colloquium in Barcelona. April 6-8, 2006. Tsoulas, G. 2007. On the grammar of number and mass terms in Greek. Retrieved 23 March 2011 from < http://thesource.dlp.mit.edu:16080/greeksynsym/papers/Tsoulas.pdf> Whorf, B. L. 1956. Language, thought, and reality: Selected writings of Benjamin Lee Whorf (J. B. Carroll, Eds.). Cambridge, MA: MIT Press.27

Recently Viewed Presentations

  • Employment &amp; Payroll Laws

    Employment & Payroll Laws

    A list of acceptable documents is provided on the form. They include, but are not limited to, a passport, a birth certificate, a social security card and a driver's license. An employer may not ask for nor mandate a particular...
  • Introduction to Poetry - 5th grade

    Introduction to Poetry - 5th grade

    A poem's form is its shape or what it looks like. Poems are divided into lines. Lines are usually grouped into . ... Like a diamond in the sky. Then the traveler in the dark. ... Introduction to Poetry -...
  • Creating PPT presentations

    Creating PPT presentations

    First identified in Ontario in 1993. A "vector borne disease" - blacklegged tick (Ixodesscapularis) A "reportable disease" since 1988 - HPPA. An "emerging disease" - migrating northward over time, with climate change
  • Science and Software - George Mason University

    Science and Software - George Mason University

    ENGH . 410: Prof / Tech . Writing. One of PSYC 333, COMM 320, COMM 335. CS 101, October 2016. ... Software testers try to find faults before the faults find users. NASA's Mars lander: September 1999, crashed due to...
  • Roma Mater - Bishop Ireton High School

    Roma Mater - Bishop Ireton High School

    Rome was the site of many famous temples, but one of the most impressive is the Pantheon with its enormous open dome. It still stands today, reconsecrated and redecorated as a church. Living in Rome. There were two basic types...
  • The Message Is Clear Workplace Skills: Communication  Essential

    The Message Is Clear Workplace Skills: Communication Essential

    The Message Is ClearWorkplace Skills: Communication. Essential Questions. How do we affect others with our listening skills? What can we do as individuals to improve our speaking and listening skills?
  • Five Steps to Lean 1. 2. 3. Define

    Five Steps to Lean 1. 2. 3. Define

    Multi-skilled workers facilitate this. Kanban links customer demand to final assembly, and then to internal and external suppliers (synchronization) Changeover time should equal one takt time for final assembly operations If customers order 100 products per day, what is the...
  • Gases,Liquids &amp; Solids

    Gases,Liquids & Solids

    GASES,LIQUIDS & SOLIDS Solids Liquids Gases Fixed volume & fixed dimensions. Resist deformation (changes of shape) ... 760 mm Hg (b) Pa (PASCALS) = N m-2 = kg m-1 s-1 1 Atm. = 101.325 kPa 1 bar = 105 kPa...