CALC-2: Book of Abstracts

Plenary lectures

Jordan Zlatev (University of Lund)

Speech, gesture and depiction in two sand-drawing cultural practices

Human communication is by default polysemiotic: it involves the spontaneous combination of two or more semiotic systems, the most important ones being languagegesture and depiction. I will propose an original cognitive-semiotic framework for the analysis of polysemiosis, contrasting this with more familiar (though ambiguous) systems based on “multimodality”. Our group has developed a coding system for the analysis of polysemiotic utterances containing speech, gesture and drawing, implemented this in the ELAN video annotation software, and applied this to the systematic analysis of 23 video-recordings of sand drawing performances on Paama, Vanuatu and 20 sand stories of the Pitjantjatjara culture in Central Australia. I will describe the coding system and provide illustrative examples from the Paamese and Pitjantjatjara data, remarking on similarities and differences in the polysemiosis of the two cultural practices. 

Anežka Kuzmičová (Charles University)

Life to text, or text to life? Children’s views on story and the self

Self-referencing, i.e., experiencing narrative text as relevant to one’s life, is well known to scholars who study readers’ responses to fiction. Across frameworks and methodologies, research tends to focus on how readers recognise similarities with what is already “there” in one’s life; then their prospective or speculative thinking is sometimes explored as an extension of this initial recognition. However, this research deals with adult and young adult readers whose capacities for autobiographical reasoning are well-developed. Little is known about how self-referencing might work in children for whom the concept of own life has yet to emerge. I will present the findings of two studies with readers aged 9-12 where self-referencing was part of the research design. In the first study, participants (N = 148) were invited to describe their story-based remindings and mental images. The second study (N = 28) employed Q methodology and dealt with becoming “lost in books” more generally. Both studies show a dominant and largely autonomous role of self-referencing in the prospective and speculative (text to life) rather than similarity (life to text) directions. The findings suggest that for many children, story reading is inherently about what one could become, irrespective of perceived realism.


Dariia Andreeva (Charles University)

The linguistic representation of the COVID-19 pandemic in Czech and Russian mass media from the point of view of the cognitive linguistics

This presentation will allow us to take a glance at the representation of the COVID-19 pandemic in the mass media discourse in the Czech Republic and in Russia from the point of view of the cognitive linguistics. Without any doubt, the coronavirus pandemic has had a global impact on different spheres of modern life, and this influence has been reflected in multiple linguistic areas, including the language of mass media. In particular, the pandemic has affected disaster journalism, which has already been gaining more and more attention from researchers in liberal arts in recent years. In this presentation, it will be revealed how different linguistic devices can be applied to depict the pandemic in Czech and Russian mass media. Moreover, examples of speech strategies and tactics, used in mass media discourse for representing the pandemic in the Czech Republic and in Russia, will be shown in the presentation. The presentation contains examples, borrowed from different Czech mass media sources («iDnes», «Deník», «Blesk», «Respekt», «Lidové noviny»), and Russian mass media sources («Kommersant», «RBK», «Izvestiya», «Komsomolskaya pravda», «Argumenty i Fakty»).

Róbert Bohát (Charles University)

The Learning Metaphor Revolution: Metaphors Students Learn by in Corpus Comenius (CorCo) of Learning Metaphors

How can we “bring on a learning revolution”? A famous pedagogical TED talk suggested that “we have to change metaphors” from a mechanical to an organic model of learning. (Robinson 2010) Corpus Comenius (CorCo) is designed to document authentic educational metaphors at administrative, professional and student levels. This paper will discuss a sample of the student subcorpus of CorCo – a preliminary analysis of the first 40 responses to the Learning Metaphor Survey (LMS) by students in Grade 9 (14-15 years old). The MIPVU method was used in identifying metaphors; the results form the foundation of a Learning Metaphor Toolkit – a tool similar to the “metaphor menu” for cancer patients. (Semino et al. 2014) This could help students identify empowering metaphors they can live by and learn by. (Lakoff and Johnson 1980; Lakoff 1998) 
One of the three prompts in LMS asked students to compare learning to something else; of the 43 metaphors supplied, almost 26% used a LEARNING IS A NATURAL PROCESS metaphor (e.g. growth, photosynthesis, symbiosis, etc.). The second most frequent metaphor with 23% of the sample was LEARNING IS A JOURNEY; the third (with 14%) was LEARNING IS PLAYING. Another prompt asked students to provide a learning metaphor that empowers them. Here, 42% of the empowering metaphors were LEARNING IS A JOURNEY; the second most popular metaphor was LEARNING IS PLAYING (16%), and the third most popular metaphor choice was LEARNING IS WAR with 14%. These preliminary data seem to indicate a trend (among students) towards a preference of “organic” or “natural” metaphors, journey metaphors and metaphors of playing, with only one response hinting at a “mechanical” metaphor of learning “working” in a machine-like manner. (Some previous observations on the organic model of embodied cognition were published in Bohát 2020; 2022) How do these student preferences compare with the reality of schooling reflected in their definitions of learning (LMS prompt 1)? And does the student voice of CorCo confirm the “metaphor learning revolution” proposal “to go from … a manufacturing model… to a model that is based more on principles of agriculture”? (Robinson 2010) 

Bohát, R. (2022, in print) Metaphors We Learn By: A Probe into the Depths of the Student’s Metaphor or: Poetically dwells the student at school. Jazykovědné aktuality č. 3-4/2022. Jazykovědné sdružení České republiky, Praha. 
Bohát, R. (2020). “‘My soul knoweth right well’: The Biblical Definition of “Soul” (Hebr. nefeš, Gr. psyché) and the Epistemology of Embodied Cognition – An Ancient Source of a Modern Concept?“ In: Masłowska, E. et al. The Soul in the Axiosphere – the Axiosphere of the Soul – Anthropological and Linguistic Images of the Soul in Intercultural PerspectiveCambridge Scholars Publishing, Newcastle, UK. 
Lakoff, G., & Johnson, M. (1980). Metaphors we live by. Chicago, IL: University of Chicago Press. 
Lakoff, G. (1998). The Contemporary Theory of Metaphor. In Metaphor and Thought (pp. 202-251). Cambridge, UK: Cambridge University Press. 
MIPVU (Metaphor Identification Procedure VU University Amsterdam). (n.d.). Retrieved April 22, 2017. 
Robinson, K. (2010): “Bring on the learning revolution!” TED, February 2010. 
Semino, E. (2014, December 17). A ‘metaphor menu’ for cancer patients. Retrieved April 22, 2017. 

James Brand, Mikuláš Preininger, Adam Kříž & Martina Ceháková (Charles University)

Investigating the socio-semantic representations of L2 speakers of English

Research on conceptual representations provides a key interlink between language and cognition. One specific stream of this research has focused on quantifying the semantics of words through the large-scale collection of normative ratings along theoretically meaningful dimensions, such as concreteness (Brysbaert et al., 2014). Whilst there exists a large body of such work for different languages, the participants have largely been restricted to L1 speakers, thus there remains relatively little attention given to quantifying the representations of L2 speakers. The aims of the present study are to provide the first large-scale investigation into the differences, and the similarities that exist between L1 English, L1 Czech and L2 English speakers, in terms of their representations of words along the dimensions of gender and valence. 

Our data for the L1 English sample are taken from the Glasgow norms database (Scott et al., 2019), which contains a list of 5,550 English words. Our L1 Czech data come from the SocioLex database (Preininger et al., 2022), a list of 2,700 Czech words. Both of the L1 datasets contain normative ratings for the dimensions of gender and valence, with 1,100 translation equivalent word forms available in both datasets. We used this list of 1,100 words and collected a new set of ratings for the English words from L2 English speakers, whose L1 was Czech (data collection is still in progress). The word list also contained 122 animate nouns that had three different variants – gender unmarked (e.g. SOLDIER) and gender marked (e.g. SOLDIER (FEMALE) and SOLDIER (MALE)). Only the unmarked versions are presented in the Glasgow norms (e.g. only SOLDIER was presented), whereas in the SocioLex norms, the Czech words were marked morphologically, indicating the gender of the referent (e.g. VOJAČKA and VOJÁK). 

As data collection is still in progress (expected completion in October 2022), we do not report on any concrete results in this abstract, but our analysis plan will aim to directly compare the ratings between the three different datasets, to provide insights into the ways in which wordss may vary in terms of their conceptual representations. We hope that this novel approach may highlight the importance of considering L2 speakers in research on quantitative semantics. 

Petra Čechová (Charles University)

Phonotactic Probability Effects on Spoken and Visual Pseudoword Perception in Czech

Phonotactic probability refers to the frequency with which phonological segments and sequences of phonological segments occur in words in a given language (Vitevich & Luce, 2004). It has been shown that phonotactic probabilities of words are important in language processing and language acquisition (Jusczyk, Luce & Charles-Luce, 1994; Mattys & Jusczyk, 2001; Pitt & McQueen, 1998). For example, words with high phonotactic probability are recognized faster by native speakers in lexical decision tasks (Luce & Large, 2001) and pseudowords with high phonotactic probability are judged as more word-like by adults (Vitevitch, Luce, Charles-Luce & Kemmerer, 1997). These effects were however tested mainly on English. In this paper we present two word-likeness rating tasks conducted on Czech. 
In experiment 1, 88 native speakers of Czech listened to recordings of 40 pseudowords with varying values of phonotactic probability in random order. They were asked to judge the pseudowords based on their word-likeness on a seven-item Likert scale. A mixed-effects model revealed that phonotactic probability is a good predictor of word-likeness ratings (χ2 (1)=16.37, p=0.000052), yet there is a lot of variance in the data. We found no effect of neighborhood density. In the on-going experiment 2, the participants rate the same pseudowords presented visually. This should show us whether the effect of phonotactic probability persists even with written stimuli. 

The described experiment 1 confirms that phonotactic probability influences processing of pseudowords to a similar extent as in English. This is an important finding; since phonotactic probability serves as a factor in many psycholinguistic experiments on English, now it can be used in a similar way on Czech. The variance in the data might be caused, among other factors, by the effect of morphology (some of the pseudowords ended with possible morphemes, some did not). This might be further accentuated by the fact that the standard calculation method of phonotactic probability (Vitevitch & Luce, 2004) puts more weight to the beginnings of words than their endings. However, when a mean of standard and reversed calculation is used, the variance diminishes. 


Jusczyk, P. W., Luce, P. A., & Charles-Luce, J. (1994). Infants′ sensitivity to phonotactic patterns in the native language. Journal of Memory and Language, 33(5), 630-645. 
Luce, P. A., & Large, N. R. (2001). Phonotactics, density, and entropy in spoken word recognition. Language and Cognitive Processes, 16(5-6), 565-581. 
Mattys, S. L., & Jusczyk, P. W. (2001). Phonotactic cues for segmentation of fluent speech by infants. Cognition, 78(2), 91-121. 
Pitt, M. A., & McQueen, J. M. (1998). Is compensation for coarticulation mediated by the lexicon?. Journal of Memory and Language, 39(3), 347-370. 
Vitevitch, M. S., & Luce, P. A. (2004). A web-based interface to calculate phonotactic probability for words and nonwords in English. Behavior Research Methods, Instruments, & Computers, 36(3), 481-487. 
Vitevitch. M. S., Luce P. A., Charles-Luce J., & Kemmerer D (1997). Phonotactics and syllable stress: Implications for the processing of spoken nonsense words. Language & Speech, 40, 47–62. 

Maroš Filip (Charles University), Filip Smolík (Charles University & Czech Academy of Sciences) & Johannes Gerwien (Heidelberg University)

Word imageability and processing of sensory-based decisions

Word imageability affects a number of measures related to word memory, learning and processing. Highly imageable words are acquired faster, remembered better, and processed faster1–6. However, some aspects of word imageability are not properly understood, for example, how does it relate to recourses in the cognitive system used for sensory processing? As a first step toward a better understanding, we aim to investigate (1) whether word imageability is related to the word’s ability to facilitate decisions about sensory properties of the word referent, and (2) whether the effect is modulated by the presence or absence of a concurrently presented visual stimulus. 

We designed a study that uses a behavioral task. In each trial, we present participants with a written word and ask them to form a mental image of its referent. Words (N=120) vary in imageability ratings. The time participants are given to imagine the referent was manipulated (1, 1.5, 2 s). Subsequently participants are asked to respond to one of two randomly assigned questions: “Is it smaller than a car?”, or “Is it heavy?”. Two versions of the task are varied between subjects, one with a blank screen during the “imagining phase”, the other showing a flickering chessboard during this period (see Fig. 1). Participants were 80 Czech and 80 German students, with the word lists consisting of translation equivalents. 

We used linear mixed models to examine how response time was related to language, question, imageability, SOA and the chessboard. The results revealed significant main effects of all variables except for the chessboard, and a number of interactions, including a four-way interaction between imageability, SOA, language and presence of chessboard, and a three-way interaction between imageability, language, and question. The effects of imageability were stronger in German (see Fig. 2); imageability effects were most pronounced for the shortest SOA and decreased in the absence of the chessboard (see Fig. 3). The presence of the distractor stimulus (chessboard) appears to slow down the decay of the imageability effect with time. Overall, the results validate word imageability as a phenomenon related to sensory processing, especially to the visual domain. 

Figure 1: Scheme of a trial.
Figure 2: Interaction between imageability, language and question.
Figure 3: Interaction between imageability, SOA and distractor stimulus presence (board =
present, non_b = absent).

1. Kroll JF, Merves JS. Lexical access for concrete and abstract words. J Exp Psychol Learn Mem Cogn. 1986;12:92-107.
2. Strain E, Patterson K, Seidenberg MS. Semantic effects in single-word naming. J Exp Psychol Learn Mem Cogn. 1995;21:1140-1154.
3. Smolík F. Imageability and neighborhood density facilitate the age of word acquisition in Czech. J Speech, Lang Hear Res. 2019;62(5). doi:10.1044/2018_JSLHR-L-18-0242
4. Prado EL, Ullman MT. Can imageability help us draw the line between storage and composition? J Exp Psychol Learn Mem Cogn. 2009;35:849-866. doi:10.1037/a0015286
5. Villena-González M, López V, Rodríguez E. Orienting attention to visual or verbal/auditory imagery differentially impairs the processing of visual stimuli. Neuroimage. 2016. doi:10.1016/j.neuroimage.2016.02.013
6. Nittono H, Suehiro M, Hori T. Word imageability and N400 in an incidental memory paradigm. Int J Psychophysiol. doi:10.1016/S0167-8760(02)00002-8

Enrique Gutierrez Rubio (Palacký University Olomouc) & Petr Kos (University of South Bohemia in České Budějovice)

Conceptualization in naming: a case study on colloquial American Spanish terms for PENIS

Within an onomasiological approach to word formation, the starting point in naming is mental processing of the concept to be named, namely categorization and conceptualization. The aim of the presentation is to demonstrate the various ways a single extra-linguistic referent can be conceptually approached by metonymy and metaphor on samples of American Spanish terms for the concept PENIS. The selection of this concept for analysis is motivated by the fact that for this one particular concept we can find hundreds of different terms displaying richness of conceptualization. 

In our approach to metonymy as a means of conceptualization, the parts of the concept’s ICM (see Lakoff 1987) which the given concept shares with other members of an already existing category lead to the classification of the concept to the existing category (the process of categorization). Other parts of the concept’s ICM which are specific for the concept within the selected category serve as a source for the actual naming. This part, or parts, of the ICM thus provide(s) mental access to the whole concept by the PART OF ICM FOR WHOLE ICM metonymy (see Radden & Panther 2004: 8). This initial metonymy may be followed by other subsequent metonymies or metaphors. From this it follows that this initial metonymy is always instrumental in naming and may serve as the starting point for further conceptualization. 

The data for the analysis has been collected from the Spanish-Czech Dictionary of Americanisms (Černý 2018), a monumental work that gathers words and idioms used exclusively in American Spanish together with their Czech equivalents. The reason why this dictionary has been chosen is because it is based on more than 20 specialized dictionaries and glossaries of Americanisms, including some works devoted to informal variants of specific geographic areas. As a result of this, it provides us with a rich number of colloquial words, including hundreds of terms for PENIS used in different regions of America. 

Our presentation will comprise a more detailed description of the role of metonymy and metaphor in naming and its application on a sample of data, which will demonstrate the richness of ways a single concept may be mentally approached. 

Černý, J. (2018). Španělsko-český slovník amerikanismů. Olomouc: Univerzita Palackého v Olomouci. 
Lakoff, G. (1987). Women, Fire, and Dangerous Things: What Categories Reveal about the Mind. Chicago, London: University of Chicago Press. 
Radden, G. & Panther, K. U. (2004). Introduction: Reflections on motivation. In G. Radden & K. U. Panther (Eds.), Studies in Linguistic Motivation (pp. 1-46). Berlin, New York: Mouton de Gruyter. 

Martin Janečka and Eliška Veselá (Charles University)

Using recursive sentences in Czech speakers with aphasia and Alzheimer’s disease

Our research is inspired by the study of Bánréti et al. (2016), it is focused particularly on the use of recursive sentences. According to their results, patients with Broca’s aphasia do not use the recursive sentences, they prefer projecting themselves into specific characters, but they communicate relevant ToM content. Thus, this is a very significant finding and an identifier that can be used to distinguish Alzheimer’s disease from aphasia. 
We set ourselves several objectives: (1) to describe the agrammatic phenomena that are part of the patient’s speech; (2) to present the patient’s difficulties in realizing selected communication skills; (3) to confront the analyses data with theoretical findings. We proceed from a linguistic analysis of a spoken dialogue with a patient. The research was conducted on 6 patients with diagnosed aphasia and the analysed material is pre-existing transcripts, which were taken for the purpose of previous research. We evaluated constructions that appeared to be recursive sentences and we compare them with 6 persons with Alzheimer’s disease. 
Our results are in agreement with the study of Bánréti et al. In patients with Broca’s aphasia, there is no use of recursive sentences, while in patients with Wernicke’s aphasia the frequency of the use of recursive sentences is slightly higher and it is clearly highest in patients with transcortical motor aphasia. 
We want to focus on Theory of Mind afterwards. Communicating relevant ToM content is predicted to be more likely only in patients with Broca’s aphasia. We also expect a loss of ability to understand other people’s intentions and opinions in people with AD. 
We claim that our findings could help in better analysis of language impairment in speakers with aphasia and with Alzheimer’s disease. 

Dana Kratochvílová (Charles University)

The Spanish conditional and grounding

The paper focuses on the Spanish conditional, which is characterised by a wide range of meanings that could be described as modal, temporal and evidential (quotative). In my analysis, I draw on cognitive grammar and on Langacker’s understanding of ground(ing) and subjectivity (Langacker 1985; 1990; 1991; 1997; 1999; 2002; 2003; 2006 inter alia). I first define the individual components of the ground (i.e. the communication situation in which the speaker and the addressees find themselves) and isolate its temporal, modal and evidential elements. Then, I focus on their subjective (implicit) reflection in the meaning of the Spanish conditional. I define the Spanish conditional as a relative verb form in the broad sense, whose basic characteristic is that the verbal meaning is not directly dependent on the actual communication situation, but on an implicitly construed secondary ground that reflects only some elements of the real ground. In analyses based on corpus material, I focus on the elements of the ground that are subjectively present in particular usage types of the Spanish conditional. This approach makes it possible to define these uses precisely and to point out what they have in common and how they differ. The approach also offers scope for a cross-linguistic comparison.

Michaela Martinková & Markéta Janebová (Palacký University Olomouc)

Coming and going in Czech: deictic, or not?

If Slavic verbs code deixis, then on prefixes; according to Filipović (2010, 253), “[t]he majority of OD-/DO-verbs are used deictically,” Slobin (2004, 8) considers the Russian pri- to be “a deictic prefix on a motion verb” and according to Malá (2015, 174), its Czech cognate při- indicates “directed motion towards the deictic centre (the speaker).” Lewandowski (2014, 44), however, argues that distribution of (most of) Slavic equivalents of come and go “is related to other, non-deictic factors,” e.g. Slavic come verbs are “preferred when the speaker wishes to adopt an arrival-oriented perspective;” similarly Janda et al. (2013) argue that ARRIVE is the prototype for the radial category of pri-. This presentation first investigates whether při- is a deictic prefix by identifying Mutual correspondence (MC) between come and Czech intransitive motion verbs prefixed by při- in a bidirectional parallel corpus of subtitles (created within InterCorp [Čermák and Rosen 2012]). MC is only 26.3%, with a translation bias: 44.7% of při-verbs are translated by come, but only 11.9% of come are translated by při-verbs. While při-verbs are typically used to describe motion to a deictic centre, in many such situations they cannot be used, e.g. in those incompatible with the ARRIVAL perspective in (1) and (2). Instead, (2) calls for the imperative pojď [DEIX-walk-IMP]; arguably, po- in imperatives of verbs coding directed motion modifies the meaning of the verb so that it invites the hearer to accompany the speaker, or to move in the direction toward the speaker (Petr et al. [1986, 418], see also Biskup [2019]). The hypothesis that the imperatives of jít [walk] are deictic is then tested; pojď/te [DEIX-walk-IMP] is indeed typically translated by come, (j)di//(j)děte [walk-IMP] (64.3%) by go. However, go also appears in 58.8% of correspondences of běž/běžte [lit. run-IMP], which, too, codes human locomotion away-from-speaker. In the next step, we will use ORAL1 to investigate factors triggering the use of (j)di/(j)děte or běž/běžte; the latter, though (to the best of our knowledge) not recognized as an imperative for jít in Czech grammar books, is even more frequent of the two in ORAL1. 

(1) Can I come in? (IC:SUBTITLES:203975) 
Můžu jít dál? 
can-1sg walk-IPFV futher 

(2) Come to momma! (IC:SUBTITLES:4027439) 
Pojď k mamince! 
DEIX-walk-IMP toward mummy 


Biskup, Petr. 2019. Prepositions and Verbal Prefixes: The Case of Slavic. Amsterdam/Philadelphia: John Benjamins Publishing Company. 
Čermák, František a Alexandr Rosen. 2012. The case of InterCorp, a multilingual parallel corpus. International Journal of Corpus Linguistics, 17(3), 411–427. 
Filipović, Luna. 2010. The Importance of being a prefix: Prefixal morphology and the lexicalization of motion verbs in Serbo-Croatian. In Hasko, V. and R. Perelmutter (eds), New Approaches to Slavic Verbs of Motion. Amsterdam/Philadelphia: John Benjamins Publishing Company. 
Janda, Laura A., Endresen, Anna, Kuznetsova, Julia, Lyashevskaya, Olga, Makarova, Anastasia, Nesset, Tore, Sokolova, Svetlana. 2013. Why Russian aspectual prefixes aren’t empty: prefixes as verb classifiers. Bloomington, IN: Slavica Publishers. 
Lewandowski, Wojciech. 2014. Deictic Verbs: Typology, Thinking for Speaking and SLA. SKY Journal of Linguistics 27, 43–65. 
Malá, Markéta. 2015. Translation counterparts as indicators of the boundaries of units of meaning: A contrastive view of the position of “come V-ing” among the patterns of the verb come. In Ebeling, S. O. and H. Hasselgard (eds), Cross-Linguistic Perspectives on Verb Constructions. Cambridge Scholars Publishing. 
Petr, Jan, Komárek, Miroslav, Kořenský, Jan, and Jarmila Veselková. 1986. Mluvnice češtiny [2]. Praha: Academia. 
Slobin, D. I. 2004. The many ways to search for a frog: Linguistic typology and the expression of motion events. In S. Strömqvist S. and L. Verhoeven (eds.), Relating events in narrative: Typological and contextual perspectives. Mahwah, NJ: Lawrence Erlbaum Associates. 

Michal Láznička (Charles University)

The role of sequential iconicity in the processing of temporal adverbial clauses: An artificial language learning study

Sequential iconicity refers to a diagrammatic ordering of elements in discourse such that it reflects the order of events in reality. While contexts such as coordination only permit an iconic order (e.g. He sat down and started singing does not allow for a reverse order interpretation), this can be overridden by using temporal adverbial clauses with connectives that express the order of the clause relative to the main clause. Thus, Before he started singing, he sat down presents the events in a non-iconic order with respect to reality. However, there is a tendency even for temporal adverbial clauses to be ordered in an iconic fashion (Diessel 2008; Rezaee & Golparvar 2016). Potential processing benefits of grammatical iconicity in general and sequential iconicity in particular remains an open question. Researchers such as Ramat (1995) have hypothesized iconic structures to be easier to process in language acquisition and learning as well as in language disorders. This has been partially confirmed by de Ruiter et al. (2018), who found that children comprehend iconically ordered temporal adverbial clauses better. 

In this paper, I will discuss artificial language learning (Fedzechkina, Newport & Jaeger 2016) as a method to investigate the benefits of iconicity. A pilot study conducted with 67 participants has shown a tendency to prefer iconic clause order. Participants were presented with an equal number of iconic and non-iconic clause orders and asked to produce novel sentences. Participants produced an above-chance proportion of iconic order both with before- and after-clauses (58.05 and 63.21 % respectively). The results of two follow up experiments will be also presented. Experiment 1 will use the same design as the pilot study, i.e. a mixture shift paradigm with both types of constructions and elicitation of novel sentences. Experiment 2 will use the same stimuli in the learning phase. A picture matching task will be used in the test phase to investigate comprehension. I expect the results to be similar to the pilot study such that more iconic orders will be elicited in Experiment 1 and responses will be faster and more accurate for iconic orders in Experiment 2. 

Diessel, Holger. 2008. Iconicity of sequence: A corpus-based analysis of the positioning of temporal adverbial clauses in English. Cognitive Linguistics 19(3). 465–490. 
Fedzechkina, Maryia, Elissa L Newport & T Florian Jaeger. 2016. Miniature Artificial Language Learning as a Complement to Typological Data. In Lourdes Ortega, Andrea E. Tyler, Hae In Park & Mariko Uno (eds.), The Usage-based Study of Language Learning and Multilingualism, 211–232. Washington, DC: Georgetown University Press. 
Ramat, Anna Giacalone. 1995. Iconicity in the Grammaticalization Processes. In Raffaele Simone (ed.), Iconicity in Language, 119–139. John Benjamins. 
Rezaee, Abbas Ali & Seyyed Ehsan Golparvar. 2016. The sequencing of adverbial clauses of time in academic English: Random forest modelling. Journal of Language Modelling 4(2). 225–244. 
Ruiter, Laura E. de, Anna L. Theakston, Silke Brandt & Elena V.M. Lieven. 2018. Iconicity affects children’s comprehension of complex sentences: The role of semantics, clause order, input and individual differences. Cognition 171. 202–224.

Wei-Lun Lu (Masaryk University)

Comparing construals of viewpoint across gospels: A Multiple-Parallel-Text approach

The presentation addresses the radically conventional nature of grammar from a cognitive constructionist perspective and its influence on the rendition of religious scriptures. In this paper, I demonstrate how that characteristic of human language is revealed by the fundamental difference in construal mediated by the different sets of linguistic tools that are available to the speakers of different languages. 
In religious discourse, such characteristic of human language is evidenced in the religious scriptures in the various languages of the world, with the rendition of the gospel in different languages being a classic example. To illustrate this point, the presentation compares the narrative viewpoint structure in a selection of parallel biblical scriptures in typologically distant languages, using English and Chinese as samples. I select Luke 19:1-10 from the existing multiple versions from the two languages as illustration and compare my results with the viewpoint analysis of the same passage laid out in Yamazaki (2006). 
Methodologically, each of the samples represents a generalization made from the selected versions of Luke 19:1-10 in the two languages compared. Such method that involves multiple parallel texts from each of the languages compared ensures the representativeness of the viewpointing strategies that are found to be involved in encoding the biblical message and are, as a result, influential to the religious communities that speak that particular language. The validity of the generalization made for each of the languages ensures the methodological rigor of the cross-linguistic comparison. 
The main theme of this presentation is to demonstrate how the language-specific conventionality of human grammar creates the various irreducibly language-specific styles in which parallel religious messages (such as the Christian gospel) is communicated—although the various (teams of) translators manage to keep the main theme of gospel intact, the detail of how exactly the gospel is communicated do vary from one language to another.

Eva Pospíšilová (Charles University)

The effect of argument semantics and relative frequency on the perception of passive constructions in Czech

Complex syntactic structures, such as passive constructions or relative clauses, have been found to emerge late in language acquisition and decline fast in agrammatic speech perception. Many studies of agrammatic speech processing have focused on the perception of semantically reversible sentences with passive construction, and the results have shown that semantically irreversible passive constructions have been processed better than passive constructions that are semantically reversible (e.g. Ansell and Flowers 1982). Although there have been many atempts to explain this phenomenon solely on the syntactic basis (see for example Grodzinsky’s trace deletion hypothesis, Grodzinsky 1986), recent works using the methods of usage-based linguistics suggest that relative frequency of the passive forms can play a role, too (Diessel 2007, Gahl and Menn 2016). Both in agrammatic and grammatic speech perception studies, it has been shown that passive forms with a higher relative frequency have been processed easier than passive forms with a lower relative frequency (e.g. Lalami 1997, Gahl et al. 2003). 
Based on these findings, we have conducted an experiment to test the influence of argument semantics and relative frequency of passive forms on the perception of Czech passive construction in Czech neurotypical speakers. Based on a corpus analysis, a total of 24 verbs with high and low relative frequency of passive forms were selected, each verb was used in four sentences with different combinations of voice, relative frequency of passive form and (in)animacy of patient (examples 1 and 2). The self-paced reading method was implemented along with a task focused on naturalness judgement.  
(1) Set of sentences containing a verb with a high relative frequency of passive form: 
a. Ředitelka hodnotila na zasedání prezentaci 
b. Prezentace byla na zasedání hodnocena ředitelkou. 
c. Ředitelka hodnotila na zasedání manažerku. 
d. Manažerka byla na zasedání hodnocena ředitelkou. 

(2) Set of sentences containing a verb with a low relative frequency of passive form: 
a. Dirigentka poslouchala v šatně nahrávku. 
b. Nahrávka byla v šatně poslouchána dirigentkou. 
c. Dirigentka poslouchala v šatně zpěvačku. 
d. Zpěvačka byla v šatně poslouchána dirigentkou. 

The results have shown that sentences with active forms of verb were overall seen as more natural than those with passive constructions. The naturalness of the sentences with passive verb form was modulated by the factor of relative frequency, with sentences containing the relatively frequent passive form scoring higher than the sentences with a relatively nonfrequent passive form. 

Ansell, B. J., & Flowers, C. R. (1982). Aphasic adults’ use of heuristic and structural linguistic cues for sentence analysis. Brain and Language, 16(1), 61-72. 
Berndt, R. S., Mitchum, C. C., & Haendiges, A. N. (1996). Comprehension of reversible sentences in “agrammatism”: A meta-analysis. Cognition, 58(3), 289-308. 
Diessel, H. (2007). Frequency effects in language acquisition, language use, and diachronic change. New ideas in psychology, 25(2), 108-127. 
Gahl, S., Menn, L., Ramsberger, G., Jurafsky, D. S., Elder, E., Rewega, M., & Audrey, L. H. (2003). Syntactic frame and verb bias in aphasia: Plausibility judgments of undergoer-subject sentences. Brain and Cognition, 53(2), 223-228. 
Gahl, S., & Menn, L. (2016). Usage-based approaches to aphasia. Aphasiology, 30(11), 1361-1377. 
Grodzinsky, Y. (1986). Language deficits and the theory of syntax. Brain and language, 27(1), 135-159. 
Hudousková, A., Flanderková, E., Mertins, B., & Tomšů, K. (2014). Hypotéza vymazávání stop a její platnost v češtině u pacientů s Brocovou afázií. Slovo a slovesnost, 75(3). 
Lalami, L. (1997). Frequency in sentence comprehension. University of Southern California. 

Mikuláš Preininger, James Brand, Adam Kříž & Markéta Ceháková (Charles University)

How semantic categories structure meaning: The organisation of socio-semantic information in Czech

In the last few decades, a large body of psycholinguistic research has been trying to quantify various semantic properties of words. With this aim, various normative studies have been conducted, capturing different dimensions of meaning, e.g. concreteness (Brysbaert et al., 2014), iconicity (Winter et al., 2017) and recently aspects of social meaning such as gender and age (Preininger et al., 2022, so-called SocioLex). The aim of this paper is to introduce an additional semantic tagging to the SocioLex dataset, which demonstrates how different semantic categories of words structure the distribution of ratings in the dataset, providing novel insights into the way our cognitive system organises semantic information. 

We used the 2700-item wordlist from the SocioLex dataset, which contains normative ratings along five different dimensions (AGE, GENDER, LOCATION, POLITICAL and VALENCE). In our semantic category tagging task, participants (N = 346, Mage = 23 years, Mratings_ per_word = 22) assigned each of the given words a minimum of 1 and a maximum of 3 out of 32 different semantic categories (see appendix for list of categories), based on which category they associated the word with most. If a participant chose more than one category, they were asked to rank them based on the association strength, 1 = strongest, 3 = weakest. We then calculated a weighted proportional value based on these rankings, so that each word had a value ranging between 0 (no association strength) and 1 (unanimous association strength) for each of the 32 categories. Using these values we could then extract the categories that had the three highest values per word and additionally tag each word with a specific semantic category, based on the category with the highest proportion value, see Table 1 for examples. 

In order to better understand whether different categories of words are more likely to carry stronger socio-semantic associations, we explored how the ratings from the SocioLex norms would aggregate based on the 1st category tags from the current study. The data show that this is indeed the case (e.g., words from categories TRANSPORT and DRUGS tended to be associated with masculinity and words from category EMOTION with femininity). We hope these category tags can be used for various applications in the study of language and cognition. 

Martin Sedláček (Charles University)

Get fired or get promoted: Beneficial or adversative, and does it really matter?

The use of the English get-passive is notoriously difficult to describe regarding its evaluative meaning. On the view of “interpersonal grammar”, we identify examples varying from beneficial (get promoted) to adversative (get fired) (Carter and McCarthy, 1999). However, we also find neutral examples (get photographed) (Villalibre, 2015). While giving invaluable insights on grammatical and pragmatic levels, previous research seems to be overly reliant on simple statistical measures, such as raw frequency or i.p.m. A more complex statistically-based analysis may reveal previously unnoticed tendencies, and answer the following research questions: 
• Can we classify lexical verbs used in the get-passive as beneficial or adversative based on quantitative data, or do we need to qualitatively analyse every concordance line to understand the context as negative or positive? What counts as adversary or beneficial? 
• Research suggests preference for dynamic verbs (Alexiadou, 2005). Are stative verbs used in the construction at all? What is their frequency compared to dynamic verbs? 
Using the spoken part of British National Corpus, I employ simple collexeme analysis (SCA), developed by Gries and Stefanowitsch (e.g. Stefanowitsch and Gries, 2003, Gries, 2015 and 2019). SCA has the potential to give more nuanced results than other association measures, given that (a.) it reflects multiple variables (corpus size, the frequency of the lexeme in and outside the construction, construction frequency), and (b.) it uses precise statistics (Fischer-Yates exact test). The main benefit of the method is that it identifies the target lexeme (in this case a verb) as either attracted or repelled by the construction (get-passive), and it also quantifies its “collostructional strength”. Following a tentative analysis, preliminary results: 
a. show a prevalence of dynamic verbs (pay, marry); stative verbs (feel, remember) are also used, but are mostly repelled by get-passives; 
b. reveal the difficulty in classifying the usage of the construction as negatively or positively evaluative, since the meaning is often determined by context (Rühlemann, 2007). 
The results also show that collostructional strength may indicate the degree of productivity in the target construction (X got bored vs. X got assigned) on a cline from fixed to productive, showing the benefits of SCA and shedding light on language productivity in general. 

BNC: British National Corpus. <> 
Flach, S. (2017): {collostructions}. An R implementation for the family of collostructional methods. R package version 0.1.0. Available from <> 
R Core Team (2016): R: A language and environment for statistical computing. Vienna: R Foundation for Statistical Computing. Available from <> 

Alexiadou, A. (2011). A note on non-canonical passives: the case of the get-passive. Organizing Grammar, 13-22. New York: Mouton de Gruyter. 
Carter, R., & McCarthy, M. (1999). The English get-passive in spoken discourse: description and implications for an interpersonal grammar. English Language and Linguistics, 3, 41 – 58. 
Gries, S. (2015). More (old and new) misunderstandings of collostructional analysis: on Schmid & Küchenhoff 2013. In: Cognitive Linguistics, 26, s. 505–536. 
Gries, S. (2019). 15 years of collostructions: Some long overdue additions/corrections (to/of actually all sorts of corpus-linguistics measures). In: International Journal of Corpus Linguistics, 24(3), 385-412. 
Rühlemann, C. (2007). Lexical grammar : The GET-passive as a case in point. ICAME Journal No. 31, 111-127. 
Stefanowitsch, A. & S. Th. Gries (2003). Collostructions: Investigating the interaction of words and constructions. In: International Journal of Corpus Linguistics 8(2). 209–243. 
Villalibre, E.C. (2015). Is the get-passive really that adversative? Miscelánea: A Journal of English and American Studies, 51, 13-26

Úvod > CALC-2: Book of Abstracts