Риторические особенности академического устного дискурса: маркеры отношения и вовлеченности

Полный текст

1. Introduction

Recent developments in corpus linguistics have facilitated exploring variation in the use of linguistic features (Poos & Simpson 2002, Farr & O’Keeffe 2002). Studying how linguistic features vary across different contexts, regions, or social groups (Heine et al. 2024, Mohammed & Sanosi 2024) helps researchers gain insights into the diverse ways in which language is used and understood. It also enables them to shed light on different communities’ values, beliefs, and norms and understand the dynamic nature of language and its adaptations to new environments and influences. Linguistic variations analysis (Zbenovich et al. 2024) additionally can result in uncovering how language is used to express individual and group identities, as well as social hierarchies and power dynamics, improving intercultural and interpersonal interactions, and finally leading to more effective communication strategies.

One of the most significant innovations in the field of corpus analysis has been the development of web-based tools and software (Graesser et al. 2004, Graesser & McNamara 2011), and web-based corpora (Michigan Corpus of Upper-level Student Papers) which enabled the researchers to gather large amounts of language data from various sources quickly, to create their visualizations and data representations, and to analyze and process them more efficiently and accurately. These tools can help identify language use patterns, trends, and variations that would be difficult to detect manually.

Interactional metadiscourse features have experienced a dramatic increase in research during the past two decades as one of the linguistic features (Boginskaya 2022, Hyland & Tse 2005, Hyland 2005b). Attitude markers have increasingly been recognized as features that express emotions and help writers or speakers express their attitudes toward their claims and findings. Engagement markers are also fundamental signs of the writer’s /speakers’ recognition of their probable readers. When writing or speaking, the addressers should feel the presence of their audience, incorporate them into their arguments, focus their attention, regard them as discourse participants, and finally lead them to the correct interpretations (Hyland 2005a). Recently, researchers have shown an increased interest in exploring interactional resources of metadiscourse, particularly attitude and engagement markers, because they help text organizers not just convey information but present a reliable picture of themselves while they are acknowledging and negotiating social relations with the audience (Wazni et al. 2023, Solnyshkina et al. 2023, Hyland & Jiang 2023). In addition, the knowledge of attitude and engagement markers reveals the presence of the audience in communications and permits the language users to focus beyond merely presenting factual information and consider how their language creates interpersonal connections in academic contexts.

Despite the rich literature supporting the interactional analysis of academic texts, there has been little concentration on academic spoken English. It seems that only writers needed to convey their attitudes and emotions and engage with their audiences as discourse participants in written academic discourse but not the speakers in the spoken ones. Therefore, questions have been raised about the use of attitude markers by native and non-native English speakers to express the writer’s emotional attitude towards propositions, focusing on feelings rather than evaluating the status or reliability of information in the academic discourse. Moreover, there is increasing concern over the role of engagement markers in their speech to directly address readers, aiming to capture their attention or involve them as active participants in the discourse.

2.   Literature review

In recent years, there has been an increasing amount of efforts in literature to theorize attitude and engagement markers of interactional metadiscourse resources. Hyland (2005a) believed that using attitude markers, one of the stance features of interpersonal elements in academic genres, is a significant authorial strategy in argumentative and evaluation-loaded texts to designate evaluation and share this evaluation with the immediate audience in an interactional dimension. He stated that the interactional resources try to involve the audience in the arguments by notifying the audience of the author’s views. The authors skillfully comment on the significance, relevance, or difficulty of an idea in the content of a text and pursue the readers’ agreement. Fundamentally, attitude markers are words (i.e. agree, surprisingly, significantly, only, important, issue, need) that aid writers to express their evaluation, feeling, and attitudes concerning the discussion in the text. In addition, these markers report to the readers about the author’s point of view and his position in the text (Hyland 2005a).

Adel (2006) also described attitude markers as “the importance of something, the interest of something, its appropriateness, and the personal emotional concomitants of linguistic material” (p.174). Hyland and Tse (2005) stated that the writers use attitude markers in their texts to indicate a position and take a stance. They use these markers to formulate evaluation, make readers agree with their points of view, and pull the readers into a scheme of agreement.

On the other hand, engagement markers play an alignment function determining the ways writers or speakers rhetorically distinguish the presence of their readers to actively pull them along with the argument, include them as discourse participants, and guide them to interpretations (Hyland 2005a). He stated that engagement markers help to bring readers into the text, guide readers to do an action or to recognize things in a way determined by the writer, hold the readers’ attention and motivate them to consider a vague issue with the writer, and place readers within apparently naturalized boundaries of disciplinary understandings, and permit writers to address readers directly by briefly interrupting the argument to offer a comment on what has been said (Hyland 2005b).

In recent years, there has been an increasing amount of literature on the use of attitude and engagement markers in different types of the texts. Ayad (2022) addressed the increasing interest in how authors engage their readers through Facebook, a prominent platform for users to share personal experiences, perspectives, comments, and feelings. The researcher used the framework by Hyland (2005a) and investigated the impact of gender and age on the expression of engagement markers in 1500 English-written Facebook status updates by students and academic staff from Egypt. Using AntConc (version 3.5.8), the freeware corpus analytic toolkit, the study found weak distinctions in age and gender regarding the frequency and type of engagement features. Analyzing stance features in academic writing also allowed for a survey of its interactional and persuasive nature. A corpus-based analysis based on the comparison of M.A. and Ph.D. theses/dissertations in applied linguistics written by the same Chinese mainland writers in English revealed developmental progression in stance-making from M.A. to Ph.D. level. It indicated Ph.D. students exhibiting more advanced use of attitude markers and an amplified ability to declare positions and involve readers (Wu & Paltridge 2021).

Previous studies have reported Arabic writers employing more attitude markers and less engagement markers than their English colleagues (Alghazo et al. 2021). The interpersonal aspect of EAP writing, linguistic background, and cultural contexts were also found to be influential on engagement marker’s choices in the research article written by Persian writers (Khatibi & Esfandiari 2021). Furthermore, the rhetorical contexts of register, genre, and disciplinary content resulted in more prevalence of stance than engagement markers in presentations from six disciplines (Qiu & Jiang 2021). Moreover, L1 Thai speakers of English indicated more attitude markers in argumentative texts than native English speakers and demonstrated differences in engagement markers, often employing reader references, directives, questions, appeals to shared knowledge, and personal asides. Although the overall stance bundles were higher in Thai, their usage exhibited less variety compared to native English speakers despite similarities in structural patterns (Papangkorn & Phoocharoensil 2021).

Opinion articles analysis from English and Arabic newspapers for stance and engagement markers and their persuasive role in constructing successful arguments demonstrated how their understanding can enhance awareness of intercultural communication in academic and journalistic contexts. It indicated that engagement were less frequently used in Arabic articles and revealed that stance markers were employed significantly more than engagement markers over both kinds of articles, with the most frequently used stance marker being self-mentions (Al-Rickaby 2020). An investigation of stance and engagement markers in argumentative essays written by EFL learners of varying writing quality based on Hyland’s (2001) framework indicated no direct correlation between the presence of engagement markers and overall essay quality (Shahriari & Shadloo 2019).

The analyses of attitude markers in different types of texts have also attracted researchers’ attention in recent years (Nayernia 2019). Azar and Hashim’s (2019) analysis of review articles for identification of frequent attitude markers and their functions exhibited more attitude markers in the conclusion section. It also showed attitudinal adjectives and adverbs being the most common types, used mainly for evaluating research and giving suggestions. The study of business emails written in English by Spanish and Chinese managers additionally demonstrated that authors with different native languages employed distinct attitude, engagement, and communication strategies based on their language and culture when using English as a common language (Carrió-Pastor 2019). However, the motivation for this study arose from the lack of corpus-based research on academic spoken English. The main objective of this research was to reveal how the use of rhetorical features is influenced by the native language of academic English speakers. Therefore, it quantitatively investigated the use of attitude and engagement markers in the speech provided by the Michigan Corpus of Academic Spoken English (MICASE) (Simpson et al. 1999) and specifically addressed the following research questions.

1. Are there any differences in the use of attitude and engagement markers across academic divisions in academic spoken English of native and non-native speakers?
2. Is there any variation in the use of attitude and engagement markers across levels of interactivity in academic spoken English of native and non-native speakers?
3. Do gender differences impact the use of attitude and engagement markers in academic spoken English of native and non-native speakers?
4. How do the native and non-native speakers of various academic roles use attitude and engagement markers in their academic spoken English?

3.   Methodology

It was necessary to follow a corpus-based analysis, which was both fourfold and quantitative, to examine the use of attitude and engagement markers by two groups of speakers across four academic divisions, five levels of interactivity, two genders, and two academic roles in MICASE. These markers were selected from the taxonomy of metadiscourse features suggested by Hyland (2004), who stated that authorial markers are indicative of the writer’s emotional attitude towards propositions rather than their epistemic stance and express various emotions such as surprise, agreement, importance, obligation, frustration. The language users typically signal these attitude markers metadiscoursively using attitude verbs (e.g., agree, prefer), sentence adverbs (e.g., unfortunately, hopefully), and adjectives (e.g., appropriate, logical, remarkable), of which the list can be found in Table 1.

Table 1. List of attitude markers investigated in MICASE 

Admittedly, I agree, amazingly, appropriately, correctly, curiously, disappointing, disagree, even, fortunately, have to, hopefully, important, importantly, interest, interestingly, prefer, pleased, must, ought, prefer, remarkable, surprisingly, unfortunate, unfortunately, unusually, understandably

Hyland (2005a) also defined engagement markers as devices that obviously address readers to focus their attention or include them as discourse participants. Table 2 lists the markers considered in this study.

 Table 2. List of engagement markers investigated in MICASE 

Incidentally, by the way, let us, let’s, ours, our, us, we, you, your, one’s

In this study, attitude and engagement markers analysis was performed on native and non-native speakers’ speeches presented by MICASE. The native group examined in this study are American English speakers and non-native speakers have backgrounds in different languages (Table 4). Table 3 shows the language status of the two speakers’ groups, the number of speakers, and the word count of MICASE.

 Table 3. Speaker and word counts by speaker categories in MICASE 

	Language Status
Speaker category	Native Speakers	Non-native speakers
Total Speakers	1,449	122
Total Words	1,493,586	201,954
% of total corpus	88%	12%

 Table 4 alphabetically shows the first language of non-native speakers in MICASE. 

Table 4. The list of the first languages of non-natives speakers in MICASE 

Arabic, Armenian, Cantonese, Czech, Dutch, Estonian. Farsi, French, German, Gujarati, Hebrew, Hindi, Hungarian, Ibo, Indonesian, Italian, Japanese, Kannada, Korean, Macedonian, Mandarin, Marathi, Polish, Portuguese, Russian, South African English, Slovak, Spanish, Swahili, Swedish, Tagalog, Telegu, Thai, Turkish, British English, Ukrainian, Urdu, Ukrainian, Unknown

Table 5 indicates the corpora of each academic division—Biomedical and Health Science, Arts and Humanities, Physical Sciences and Engineering, and Social Sciences and Education—available in MICASE and analyzed in this research. 

Table 5. Speaker and word counts by academic division in MICASE

Academic Division	Humanities & Arts	Social Sciences & Education	Biological & Health Sciences	Physical Sciences & Engineering
Speech Events	36	35	32	36
Speakers	349	452	257	314
Words	434,669	420,347	325,456	358,776
% of Total Corpus	26	25	19	21
% Male	56	37	41	55
% Female	44	63	59	45
% Faculty	63	44	55	44
% Students	29	55	42	52

Table 6 indicates different levels of interactivity or discourse modes in MICASE. This study investigated highly interactive, mostly interactive, highly monologic, mostly monologic, and mixed modes in the use of attitude and engagement markers by native and non-native speakers. 

Table 6. Speaker and word counts by levels of interactivity in MICASE

PrimaryDiscourseMode	Monologic	Panel	Interactive	Mixed	Totals
Speech Events	61	9	57	25	152
Speakers	472	133	643	323	1,571
Words	554,335	141,505	715,333	284,367	1,695,540
% of Total Corpus	33	8	42	17
% Male	50	27	46	51
% Female	50	73	54	49
% Faculty	84	16	26	54
% Students	14	76	63	39

Table 7 indicates the number of female and male speakers and their total words included in MICASE. 

Table 7. Speaker and word counts by speaker categories 

Speaker Category	Gender
	Male	Female
Total Speakers	729	842
% of Total Words	786,487	909,053
Total Corpus	46%	54%

Table 8 shows the statistics of the academic roles investigated for their engagement and attitude markers in this study. 

Table 8. The academic role of the people in MICASE 

Academic Role		Total Speakers	Total Words	% of Total Corpus
Faculty		160	825,829	49%
Students	Undergraduates	782	368,433	22%
	Graduates	257	373,915	22%

Instruments

The Michigan Corpus of Academic Spoken English (MICASE) is readily available without any restrictions at (ENA, August 31, 2025)¹ (Simpson et al. 1999). It contains transcriptions of almost 1.7 million words of academic spoken English, totaling 200 hours of recordings. The creators of this valuable database aimed to track general changes in speech patterns as individuals gain experience in university culture. While we have extensive knowledge about how academic writing develops as students progress, our understanding of changes in spoken language within academic cultures remains limited. MICASE specifically focused on the prevalent speech patterns within the University of Michigan in Ann Arbor. The corpus includes speakers from various roles, such as faculty, staff, and students at all academic levels, as shown in Table 8. The creators hope this rich resource will aid developing more effective materials for teaching and testing English as a Second Language and English for Academic Purposes and will help explore the integration of corpus-based approaches into these programs.

Data collection method

This study was based on the data provided by MICASE. To answer the first research question, implying the differences between native and non-native speakers across academic divisions, the researchers filtered the corpus by each academic division and searched the attitude (Table 1) and engagement markers (Table 2) in the speech of native and non-native speakers.

Next, all of the attitude (Table 1) and engagement markers (Table 2) were separately searched for across levels of interactivity, including highly interactive, mostly interactive, highly monologic, mostly monologic, and mixed (Table 6) (addressing the second research question), across genders (Table 7) (the third research question), and across academic roles (Table 8) to find the differences between native and non-native speakers (the last research question).

Data analysis

This study needed the extraction of the frequency counts provided by MICASE into SPSS software for the use of attitude and engagement markers by native and non-native speakers across academic divisions, levels of interactivity, genders, and academic roles. Because the word counts were not equal across the corpora, these frequency counts were reported by per 1000 words. Then, the descriptive statistics, including frequency, mean, and standard deviation, was computed. To indicate the degree of significance or non-significance of the differences between the two groups of speakers across academic divisions, levels of interactivity, genders, and academic roles (independent variables) in attitude and engagement markers (dependent variables) utilization, the researchers used the UNIANOVA (One-way analysis of variance) inferential test.

4.   Results

This study intended to investigate a corpus of academic spoken English based on the adopted taxonomy to explore, compare, and contrast native and non-native speakers across academic divisions, levels of interactivity, genders, and academic roles in the use of attitude and engagement markers. To answer each research question, it represented the descriptive statistics, mean and standard deviation, and inferential statistics of UNIANOVA.

To answer the first research question concerning the differences between native and non-native English speakers across academic divisions, this study computed descriptive statistics for 23 attitude markers and 10 engagement markers (Table 9).

Table 9. The descriptive statistics 

		Attitude markers			Engagement markers
Academic Divisions	Language status	Mean	Std. Deviation	N	Mean	Std. Deviation	N
Biological and Health Sciences	Native speakers	23	56.3	23	872	2210.3	10
	Non-native speakers	1.8	5.40	23	54.2	104.2	10
Humanities and arts	Native speakers	31.2	96.4	23	1187.3	3118.3	10
	Non-native speakers	1.3	3.6	23	28.4	53.5	10
Physical Sciences and Engineering	Native speakers	17.3	49.2	23	1061.5	2740.1	10
	Non-native speakers	1.3	4.4	23	122.7	251.5	10
Social Sciences and Education	Native speakers	39	99.8	23	1144.2	2925.2	10
	Non-native speakers	1.9	5.8	23	97.7	193.3	10

According to Table 9, the native speakers’ means of attitude and engagement markers were higher than the non-natives in all four academic divisions. To examine the degree of significance of these differences between the groups across these academic divisions using the inferential test of UNIANOVA (Table 10) was necessary.

According to Table 9, the native speakers’ means of attitude and engagement markers were higher than the non-natives in all four academic divisions. To examine the degree of significance of these differences between the groups across these academic divisions using the inferential test of UNIANOVA (Table 10) was necessary.

Table 10. The UNIANOVA inferential test 

	Source	Type III Sum of Squares	df	Mean Square	F	Sig.*	Partial Eta Squared	Noncent. Parameter	Observed Power
Attitude markers	Corrected Model	37473.125	7	5353.304	1.715	.108	.064	12.002	.690
	Intercept	39414.397	1	39414.397	12.624	.000	.067	12.624	.942
	Academic Divisions	3226.364	3	1075.455	.344	.793	.006	1.033	.116
	Language status	31226.136	1	31226.136	10.002	.002*	.054	10.002	.882
	Academic Divisions Language status	3020.625	3	1006.875	.322	.809	.005	.967	.112
	Error	549491.478	176	3122.111
	Total	626379.000	184
Engagement markers	Corrected Model	20260579.600a	7	2894368.514	.752	.629	.068	5.264	.301
	Intercept	26083280.000	1	26083280.000	6.777	.011	.086	6.777	.729
	Academic Divisions	318810.900	3	106270.300	.028	.994	.001	.083	.055
	Language status	19621805.000	1	19621805.000	5.098	.027*	.066	5.098	.606
	Academic Divisions Language status	319963.700	3	106654.567	.028	.994	.001	.083	.055
	Error	277106086.400	72	3848695.644
	Total	323449946.000	80

*p 0.05

According to Table 10, the difference in the two groups’ frequency of attitude and engagement markers in academic divisions was significant (p=0.002 and F=10.002) and (p=0.027 and F=5.098), respectively. The value of eta squared was equal to 0.054 for attitude and 0.066 for engagement markers; therefore, the academic divisions accounted for almost 4.5% and 6.6% of the frequency of attitude and engagement markers. In other words, there was a significant difference between the two groups of native and non-native speakers in using the markers across all academic divisions.

To compare the native and non-native groups in the attitude and engagement markers use across levels of interactivity (the second research question), this study used the descriptive (Table 11) and inferential statistics (Table 12).

According to Table 11, the native speakers’ means of attitude and engagement markers were higher than those of the non-natives across all of the levels of interactivity. The inferential test of UNIANOVA of which the results were indicated in Table 12 was used to indicate the degree of significance of these differences.

Table 11. The descriptive statistics 

Level of interactivity	Language status	Attitude markers			Engagement markers
		Mean	Std. Deviation	N	Mean	Std. Deviation	N
Highly interactive	Native speakers	38.2609	110.63339	23	1932.7000	5106.74400	10
	Non-native speakers	.6087	1.61637	23	53.0000	107.68266	10
Highly monologic	Native speakers	11.1739	23.53653	23	306.2000	749.33493	10
	Non-native speakers	.0000	.00000	23	.0000	.00000	10
Mostly monologic	Native speakers	28.5217	80.59995	23	1037.7000	2692.13253	10
	Non-native speakers	2.0000	5.68091	23	70.0000	140.00159	10
Mostly interactive	Native speakers	26.4783	70.37489	23	852.0000	2149.29916	10
	Non-native speakers	2.1304	6.19607	23	95.1000	199.99747	10
Mixed	Native speakers	17.3043	46.08237	23	659.3000	1665.14204	10
	Non-native speakers	1.7391	5.52035	23	97.7000	180.21965	10
	Non-native speakers	1.2957	4.55372	115	63.1600	142.61341	50

 Table 12. The UNIANOVA inferential test 

		Source	Type III Sum of Squares	df	Mean Square	F	Sig*.	Partial Eta Squared	Noncent. Parameter	Observed Power
Attitude markers		Corrected Model	40727.083	9	4525.231	1.710	.088	.065	15.386	.773
		Intercept	37811.309	1	37811.309	14.285	.000	.061	14.285	.964
		Level of interactivity	5295.148	4	1323.787	.500	.736	.009	2.000	.169
		Language status	30555.657	1	30555.657	11.543	.001*	.050	11.543	.923
		Level of interactivity Language status	4876.278	4	1219.070	.461	.765	.008	1.842	.158
		Error	582340.609	220	2647.003
		Total	660879.000	230
Engagement markers		Corrected Total	623067.691	229
		Corrected Model	34880124.410a	9	3875569.379	.936	.498	.086	8.428	.437
		Intercept	26047753.690	1	26047753.690	6.294	.014	.065	6.294	.699
		Level of interactivity	7621294.460	4	1905323.615	.460	.765	.020	1.842	.154
		Language status	19999678.410	1	19999678.410	4.833	.030*	.051	4.833	.585
		Level of interactivity Language status	7259151.540	4	1814787.885	.439	.780	.019	1.754	.148
		Error	372453962.900	90	4138377.366
		Total	433381841.000	100
	*p 0.01

Based on Table 12, the differences between levels of interactivity by two groups were significant (p = 0.001 and F = 11.543) and (p = 0.030 and F = 4.833) for both attitude and engagement markers respectively. The squared value of eta was equal to 0.050 for the former and 0.051 for the latter; therefore, almost 5% and 1.5% of the changes in the frequencies of attitude and engagement markers were accounted for by the levels of interactivity. In other words, native speakers used more attitude and engagement markers than non-native speakers in highly interactive, highly monologic, mostly monologic, mostly interactive, and mixed academic spoken English in MICASE.

To answer the third research question, focusing on differences between native and non-native speakers’ use of attitude and engagement markers across two genders in academic spoken English, this study made use of descriptive (Table 13) and inferential statistics (Table 14).

Table 13. The descriptive statistics 

Gender	Language status	Attitude markers			Engagement markers
		Mean	Std. Deviation	N	Mean	Std. Deviation	N
Female	Native speakers	72.6087	198.88959	23	2717.8000	6996.23043	10
	Non-native speakers	4.0000	10.95445	23	163.9000	317.16398	10
Male	Native speakers	49.0870	128.31138	23	2068.3000	5360.61160	10
	Non-native speakers	2.4783	7.26054	23	151.9000	309.68854	10
	Non-native speakers	3.2391	9.22119	46	157.9000	305.15051	20

According to Table 13, the native speakers’ means of attitude and engagement markers were more than the non-natives across the two genders.

According to Table 14, the difference in the frequency of attitude and engagement markers across genders by two groups was significant (p=0.022 and F=5.433) and (p=0.048 and F=3.566) respectively. The values of eta squared were equal to 0.058 and 0.067 for attitude and engagement markers; therefore, almost 5.8% and 6.7% of the changes in scores were accounted for by the gender variable. That is, female native speakers significantly employed more attitude and engagement markers than female non-native speakers and male native speakers significantly employed more attitude and engagement markers than male non-native speakers.

 Table 14. The UNIANOVA inferential test 

		Source	Type III Sum of Squares	df	Mean Square	F	Sig.*	Partial Eta Squared	Noncent. Parameter	Observed Power
Attitude markers		Corrected Model	82720.783	3	27573.594	1.963	.125	.063	5.888	.490
		Intercept	94464.174	1	94464.174	6.724	.011	.071	6.724	.727
		Gender	3606.261	1	3606.261	.257	.614	.003	.257	.079
		Language status	76331.522	1	76331.522	5.433	.02**	.058	5.433	.635
		Gender Language status	2783.000	1	2783.000	.198	.657	.002	.198	.072
		Error	1236259.043	88	14048.398
		Total	1413444.000	92
Engagement markers		Corrected Model	52068926.475a	3	17356308.825	.891	.455	.069	2.674	.225
		Intercept	65073459.025	1	65073459.025	3.342	.076	.085	3.342	.428
		Gender	1093955.625	1	1093955.625	.056	.814	.002	.056	.056
		Language status	49958955.225	1	49958955.225	3.566	.048**	.067	3.566	.505
		Gender Language status	1016015.625	1	1016015.625	.052	.821	.001	.052	.056
		Error	700919071.500	36	19469974.208
		Total	818061457.000	40
	*p 0.05

To answer the last research question, investigating the differences between native and non-native speakers’ use of attitude and engagement markers across four academic roles, this study made use of the descriptive indicated in Table 15. 

Table 15. The descriptive statistics 

Academic Role	Language status	Attitude Markers			Engagement markers
		Mean	Std. Deviation	N	Mean	Std. Deviation	N
Faculty	Native speakers	62.0435	158.40411	23	2388.8000	6117.91329	10
	Non-native speakers	3.4783	10.70204	23	170.7000	346.27993	10
Graduate	Native speakers	25.3478	66.33428	23	925.3000	2413.25566	10
	Non-native speakers	2.3043	5.61196	23	118.5000	230.47740	10
Other	Native speakers	8.0000	24.27869	23	475.1000	1286.90882	10
	Non-native speakers	.2609	1.05388	23	13.2000	25.81042	10
Undergraduate	Native speakers	26.3478	82.77604	23	998.7000	2544.72435	10
	Non-native speakers	.3913	1.30520	23	13.4000	24.81129	10

According to Table 15, the native speakers’ means of attitude and engagement markers were more than that of the non-natives in all four academic roles. However, indication of the degree of significance of these differences needed the inferential test of UNIANOVA (Table 16).

Table 16. The UNIANOVA inferential test 

		Source	Type III Sum of Squares	df	Mean Square	F	Sig.*	Partial Eta Squared	Noncent. Parameter	Observed Power
Attitude markers		Corrected Model	73926.000	7	10560.857	2.278	.030	.083	15.949	.830
		Intercept	47232.087	1	47232.087	10.190	.002	.055	10.190	.888
		Academic Role	19939.000	3	6646.333	1.434	.235	.024	4.302	.376
		Language status	38223.391	1	38223.391	8.246	.005*	.045	8.246	.815
		Academic Role Language status	15763.609	3	5254.536	1.134	.337	.019	3.401	.302
		Error	815809.913	176	4635.284
		Total	936968.000	184
Engagement markers		Corrected Model	45732325.587a	7	6533189.370	1.014	.429	.090	7.096	.407
		Intercept	32559692.113	1	32559692.113	5.052	.028	.066	5.052	.602
		Academic Role	11957017.838	3	3985672.613	.618	.605	.025	1.855	.173
		Language status	24999598.013	1	24999598.013	3.899	.04*	.069	3.899	.513
		Academic Role Language status	8775709.738	3	2925236.579	.454	.715	.019	1.362	.137
		Error	464028603.300	72	6444841.713
		Total	542320621.000	80
	*p 0.05

According to Table 16, the difference in the frequency of attitude and engagement markers across four academic roles by two groups was significant (p=0.005 and F=8.246; p=0.041 and F=3.899) for attitude and engagement markers respectively. The values of eta squared were equal to 0.045 and 0.069; therefore, almost 4.5% for attitude and 6.9% for engagement markers of the changes in scores were accounted for by the four academic roles.

In other words, two groups of native speakers and non-native speakers differed in making use of attitude and engagement markers across four academic roles. That is, faculty native speakers significantly employed more attitude and engagement markers than faculty non-native speakers. Graduate native speakers significantly employed more attitude and engagement markers than graduate non-native speakers. Undergraduate native speakers significantly employed more attitude and engagement markers than undergraduate non-native speakers. Native speakers of other academic roles made use of attitude and engagement markers more than non-native speakers of other academic roles in MICASE.

5.   Discussion

The present study firstly intended to determine the differences between native and non-native speakers of English in the use of attitude and engagement markers across academic divisions in the corpus of academic spoken English namely MICASE. The results (Table 10) indicated that native speakers of the soft sciences scored higher in attitude markers because designating evaluation and sharing this evaluation with the immediate audience in an interactive dimension is one of the stance features of interpersonal elements in academic genres and a significant authorial strategy in argumentative and evaluation-loaded texts. By informing the audience of the speakers’ opinions, it gives the speakers the opportunity to engage them in the discussion. According to Hyland (2005a), it also helps people comment expertly on the importance, relevance, or difficulty of the ideas of a text and seek the listeners’ approval. It can also be linked to native speakers’ willingness to express an opinion and take a stand in the soft sciences. In soft sciences more so than in hard sciences, they employ the attitude markers to develop judgments, persuade listeners to agree with their points of view, and draw the audience into a scheme of agreement, according to Hyland and Tse (2005).

Soft sciences tend to use engagement markers more frequently in their academic writing as they try to actively engage their audience, incorporate them as participants in the discourse, and direct their interpretations (Hyland 2005a). Moreover, these sciences are inclined to involve their readers in the speech, guide them to take action or perceive things from the speaker’s perspective, maintain their attention, and motivate them to explore ambiguous issues together with the speaker, all while placing them within the disciplinary boundaries. This contrasts with a study by Back (2014), which suggested that non-native speakers tend to display a higher degree of subjectivity and personality by overusing both attitude and engagement markers in research articles.

However, non-native speakers of English not only show lower levels of attitude and engagement markers usage but also use these markers similarly across different academic divisions. This can be attributed to their lack of familiarity with expressing attitudes, involving their audience, addressing objections, and leading them to specific interpretations in English as their second language (Hyland 2005b). This finding aligns with a study by Alghazo et al. (2021), which revealed that Arabic academic writers use engagement markers less frequently than English academic writers.

Hence, this study disagreed with Khatibi and Esfandiari’s (2021) suggestion that writers’ linguistic background and cultural contexts solely determine rhetorical patterns in their research articles, as non-native group in MICASE did not differentiate their usage of engagement markers across academic divisions. Instead, we attribute their lower rate of engagement and use of attitude markers to their lack of proficiency or experience in English rhetorical strategies. Overall, native speakers’ employment of attitude and engagement markers appears to be influenced by the rhetorical contexts of register, genre, and disciplinary content.

Secondly, the current analysis suggested that native speakers consistently used a higher frequency of attitude and engagement markers compared to non-native speakers across various levels of interactivity (Table 12), such as highly interactive, mostly monologic, mostly interactive, mixed, and highly monologic. This discrepancy can be attributed to the interactive nature of the discourse, allowing native English speakers to skillfully employ listeners’ references, directives, questions, appeals to shared knowledge, and personal asides. Additionally, the highly monologic level of interactivity did not require speakers to express their attitudes or actively engage the audience throughout the speech, unlike the interactive level, where both parties need to emphasize importance, constraints, and gaps, compare and contrast ideas, and evaluate the presented points of view to advance the argument.

The study suggested that native speakers in interactive modes tend to highlight positive aspects of negotiation, positively evaluate content, and express assessment and significance more frequently through an increased use of attitude markers. Furthermore, they utilize engagement markers to directly address listeners and provide instructions on actions or preferences. Lastly, this rhetorical strategy is also employed by native speakers to make the audience relate to familiar concepts and involve them by reiterating the truthfulness of the speakers’ statements more often than non-native speakers in academic spoken English.

Regarding the third research question, implying the differences between the two groups in the use of attitude and engagement marker across two genders, the results (Table 14) contradicted Ayad’s (2022) findings, which suggested weak gender distinctions in the type and frequency of engagement markers. We found that that female native speakers actively engage their listeners in the discourse by anticipating their concerns, expectations, or objections more than males or even non-native speakers. Female native speakers actively involve their listeners in negotiations, interact with them, and prompt responses. On the other hand, male native speakers seem to be less concerned than females about representing dialogues with their audience, making them perform specific cognitive acts, arousing their interest with rhetorical questions, and establishing a relationship with them, leading to a lower use of engagement markers. However, compared to their non-native male counterparts, male native speakers exhibit more concern about these aspects.

The higher rate of attitude markers usage by female native speakers indicates their expression of affective attitudes towards information, conveying emotions like surprise, importance, obligation, and agreement. It also reflects their greater expression of assessment, significance, and position on certain issues to the listeners, emphasizing the listeners’ importance and prompting them to take specific actions in complex academic discussions.

Concerning the fourth research question, focusing on the differences between the native and non-native speakers in the use of attitude and engagement markers across academic roles, this study demonstrated native faculties’ increased expression of their evaluations, emotions, and attitudes during their speech discussions (Table 16). They utilized attitude markers to present their viewpoints and positions to the listeners (Hyland 2005a). Native faculties employed these markers more frequently than individuals in other academic roles because they needed to assert a stable position and encourage the audience’s agreement, making it challenging to question their views or opinions. Native faculties were more effective in influencing the listeners’ responses, urging them to provide solutions or answer questions posed during their presentations, as reflected in their higher use of attitude markers. This enabled the native faculties to emphasize the significance of their research area, validate their expertise, highlight the originality of their perspectives, point out gaps in research development, and evaluate previous works in related fields. As explained by Hyland (2005a), attitude markers help researchers "create a research space for engineers, assert their learned authority and expertise, solicit readers’ acceptance of claims, and reach consensus." Furthermore, native faculties employed engagement markers to make the listeners recognize familiar concepts, engage them, and build relationships to ensure their attentiveness and understanding during the presentations.

In contrast to the findings of Wu and Paltridge (2021), which showed progress in stance-making from M.A. theses to Ph.D. dissertations of students, this study indicated that graduate and undergraduate students did not significantly differ in both attitude and engagement markers usage. This finding also contradicted Crosthwaite et al. (2017), who found that professional reports displayed a narrower set of linguistic devices compared to student writers, who tended to use a wider range of stance feature types in discussing both others’ and their own personal stance, across whole texts and by section. The study suggests that native faculties established a stronger connection with listeners, essentially instructing them to take specific actions, through their higher use of engagement markers.

The study has some limitations. The scope of this study was limited two rhetorical features in MICASE and other metadiscourse features or other corpora data were not considered here. In addition, this study presented only a quantitative analysis and a qualitative one was not done to exactly consider the occurrence of these features or control their functions. It does not engage with the age or the intra-analysis of academic divisions in MICASEand was based on four variables and did not consider the speech event type, participant level, or first language of the speakers as other variables determined in MICASE. Another potential problem is that this study only considered two groups of language users, North American English speakers and non-native speakers. It did not focus on the other groups like near-native speakers, native speakers of non-American English, or unknown ones.

6.   Conclusion

This research focused on how native and non-native English speakers utilize attitude and engagement markers in academic spoken English. It analyzed MICASE to determine whether these language users differed in their use of these rhetorical features across academic divisions, levels of interactivity, genders, and academic roles.

The results supported the ideas that native speakers’ epistemology and research practices within the discourse community influenced the frequency patterns of attitude and engagement markers in their speech and non-native speakers were aware of the need to adhere to disciplinary speaking standards. Additionally, they indicated sensitivity of native speakers to levels of interactivity with their higher use of attitude and engagement markers, indicating a greater awareness of their audience and the purposes of the interaction. The results showed cultural backgrounds influence communication styles; non-native speakers are faced with challenges in expressing interpersonal stances in English; finally, situational factors play a really influential role in communication. This study also highlighted the interpersonal dynamics of academic discourse and how language functions to establish relationships, manage interactions, and convey speaker attitudes. It showcased linguistic choices reflect cultural norms and expectations, which is crucial for understanding communication in diverse academic environments.

A notable finding was the gender-specific use of these markers, with female and male academics employing different strategies to varying extents, resulting in distinct interactive effects. It clarified the influence of gender upon communication styles in academic settings. It highlighted the active role of female speakers in engaging their audience, which can inform future studies on gendered communication. Moreover, faculty native speakers used attitude and engagement markers more frequently to construct persuasive arguments during interactions compared to individuals in other academic roles. This aspect of our study illustrated the effect of authority and expertise on language choices. It empowered our understanding of register, i.e., different speaker roles necessitate varied rhetorical approaches within academic genres.

The study concluded that speakers with different mother tongues, genders, and academic roles used various attitudinal and engagement strategies in English as a lingua franca. While disciplinary community and cultural background played a role in shaping speaker positioning, other factors such as personality differences, stylistic preferences, previous education, and supervisors’ feedback also influenced the speakers’ use of attitude and engagement markers. Additionally, the research supported the notion that the use of these markers is a form of social commitment, linked to the norms and expectations of specific cultural and professional communities, and influenced by particular settings and contexts.

The findings can be integrated into instructional materials to guide students on effectively using interactive resources such as engagement and attitude markers in their presentations. Students can benefit from being aware of the appropriate use of attitude and engagement markers to effectively engage with their listeners. The use of engagement markers helps build strong speaker-listener relationships and persuasive arguments, aligning with established rules of English speech.

This research has thrown up many questions in need of further investigation. Further work needs to be done to consider the use of attitude and engagement markers by the people of different first languages, ages, and academic level and across different speech events.

 

_{¹ https://quod.lib.umich.edu/cgi/c/corpus/corpus?c=micase;page=mbrowse}

Об авторах

Аркан Альнайили

Исламский университет Азад

Email: Arkanalnayly34@gmail.com
ORCID iD: 0009-0004-5913-8159

получил степень магистра на кафедре английского языка Исламского университета Азад, филиал в Хорасгане, Исфахан, Иран. В настоящее время занимает должность директора по надзору за образованием в провинции Дивания. Его научные интересы включают дискурсивный анализ

Исфахан, Иран

Сара Мансури

Исламский университет Азад

Email: saramansouri@iau.ac.ir
ORCID iD: 0000-0002-7792-6640

имеет степень PhD в области преподавания английского языка и является доцентом кафедры английского языка Исламского университета Азад, филиал в Наджафабаде, Иран. Она специализируется в области корпусной лингвистики и дискурсивного анализа.

Наджафабад, Иран

Париваш Эсмаили

Исламский университет Азад

Автор, ответственный за переписку.
Email: parivashesmaeili@iau.ac.ir
ORCID iD: 0000-0002-4622-9009

доцент кафедры английской литературы факультета английского языка Исламского университета Азад, филиал в Наджафабаде, Иран. Ее научные интересы включают когнитивную лингвистику, когнитивную поэтику, лингвистические исследованияя на микротекстовом уровне

Наджафабад, Иран

Список литературы

Дополнительные файлы