Daria Schütz

Linnea Traulsen

1 Introduction

Linguistics is broadly defined as the “scientific study of language” (Bieswanger and Becker 2). Some of its important subfields include phonetics, phonology, morphology, syntax, semantics and pragmatics. The research at hand belongs to the subfield of pragmatics, which is defined by Bieswanger and Becker as “the systematic study of how people understand and communicate more than the literal meaning of words or sentences” (152). Discourse markers, which are parts of speech used to maintain fluency in spoken language, belong to this discipline, as they contribute significantly to all utterances made by people. Not only do discourse markers make up a large percentage of utterances in general (Schmidt et al. 97), they also significantly contribute to the meaning and interpretation of speech (Fraser 950). Since pragmatics is a subfield that also deals with the influence of culture on interpretation of speech, we asked ourselves whether the variety of English made a difference in the usage of discourse markers, specifically:

Is there a difference in frequency of the discourse marker like between Canadian and British English?

2 Literature Review

Discourse markers can serve “as pause fillers, hedges, highlighting or focus devices, as well as indicators of vagueness or ‘loose language’’’ (Tagliamonte 1897). The focus of this paper lies on the discourse marker like. Especially for the usage of like, it must be differentiated between like as a discourse marker or as a verb or preposition, e.g. “I wouldn’t like to do their job.” (no discourse marker) and “I mean that like kicks it off” (discourse marker). We define discourse markers to be part of fluencemes as “[f]luencemes can […] be realized, for example, as […] discourse markers (such as you know and like, etc.)” (Schmidt et al. 95). For the contextualization of this paper, it is critical to understand that a study conducted by Schmidt et al. “revealed a consistent fluenceme distribution across the four varieties” (Australian, British, Canadian and New Zealand English) (Schmidt et al. 94). Another study found that age was a significant factor in the usage of like amongst teenagers and children from 10 to 19 years old, with 15- to 16-year-olds using it the most and 17- to 19-year-olds the least (Tagliamonte 1903). Furthermore, Tagliamonte’s “pilot investigation reveals a concentration of these forms amongst […] female speakers in particular” (1886). In that paper, this was combined with the common assumption “that females lead linguistic change” (Tagliamonte 1886).

3 Methodology

We used two corpora belonging to the International Corpus of English (ICE), the Canadian Spoken Corpus (ICE-CAN spoken) and the Great Britain Spoken Corpus (ICE-GB spoken). The sub-corpora belonging to ICE were made to be very similar so that the comparison of the different varieties of English can be more easily accomplished. The corpora were uploaded to the program AntConc, which was used to search for the discourse marker like. We searched with a KWIC search, then exported the results as a .txt file. This file was uploaded to Excel where we randomized the order of the results and selected the first 100 hits.

Every hit was manually annotated with “yes”, “no” or “maybe” to identify them as containing a discourse marker or not. The hits tagged as “maybe” were discussed among the two researchers. This table was analyzed using R-Studio. To gain an overview, the functions ‘summary’ and ‘table’ were used, after which those results were visualized with the ‘barplot’ and ‘mosaicplot’ functions. Finally, the data was tested for statistical significance with the Chi-Square-Test.

4 Results

Our research revealed a difference between CAN-ICE and GB-ICE: In the CAN-ICE data, like was used as a discourse marker 57 times, with other usages making up 43 of the 100 hits. Comparatively, the GB-ICE hits contained only 17 uses of like as a discourse marker, with 83 comprising other usages, again coming out to a total of 100 data points. Since we worked with a sum of 100 data points for each corpus, the percentages, visualized in a bar plot (Fig.1, for another visualization as a mosaic plot, refer to the appendix.) below, are the same as the raw data presented here.

GB-ICE showed a much more striking difference in the usage of like, with only 17 percent of hits being discourse markers and 83 percent using it as either a verb or a preposition.

Figure 1: Frequency of the discourse marker like in Canadian and British English.

Using the Chi-Square-Test, a p-value of < 2.2e-16 was calculated, showing that the results are highly statistically significant, with the variety of English (Canadian or British) and the frequency of the discourse marker like being highly interdependent.

5 Discussion

The analyses showed that Canadian English speakers use the discourse marker like about three times more often than British English speakers. Even if factors like age, gender and social background were not considered for our study, the results are somewhat contrary to the findings from the papers we used as sources for our research (see Literature Review; Schmidt et al. 94). In our study, the variety of English does play a significant role in the frequency of the discourse marker like in Canadian and British English. This could be ascribed to the limited data used and the missing differentiation of other social factors. Nevertheless, this topic still constitutes an interesting field of potential further research.

6 Conclusion

Based on our research, we conclude that there is a difference in frequency of the occurrence of the discourse marker like between Canadian and British English, with Canadian English speakers using the discourse marker like about three times more often than British English speakers. Still, considering other social factors such as age, gender, or social class would have created a broader image of the reasons for our results and could have provided a more detailed overview of the topic. Further studies should consider these factors in their research, as well as a larger data set containing other varieties of English, since ours was quite limited. Another interesting component of further studies could be the consideration of other discourse markers, such as just and so, since they could provide a much more comprehensive picture of the overall usage of discourse markers in different varieties of English.

Works Cited

Bieswanger, Markus & Becker, Annette. Introduction to English Linguistics. Tübingen, Narr Francke Attempto Verlag GmbH + Co. KG, 2021

Fraser, B. “What are discourse markers?” Journal of Pragmatics, 31, 1999, 931-952

Schmidt, K., Götz, S., Jäschke, K. & Gries, S. “Same, same, but erm sort of different? Comparing three kinds of fluencemes across Australian, British, Canadian, and New Zealand English.” Research in Corpus Linguistics 13/2, 2025, 94-123

Tagliamonte, S. “So who? Like how? Just what? Discourse markers in the conversations of Young Canadians.” Journal of Pragmatics, 37, 2005, 1896-1915

Appendix

Figure 2: Frequency of the discourse marker like in Canadian and British English