English Language Teaching and Linguistics Studies ISSN 2640-9836 (Print) ISSN 2640-9844 (Online) Vol. 6, No. 2, 2024 www.scholink.org/ojs/index.php/eltls 129 Original Paper The Pragmatic Motivation of Atypical EnglishWay into/to Construction Yuting Gu1* &Ying Huang2 1 School of Foreign Studies, China University of Mining and Technology, Xuzhou, China 2 School of Foreign Studies, China University of Mining and Technology, Xuzhou, China *Yuting Gu, School of Foreign Studies, China University of Mining and Technology, Xuzhou, China Received: February 11, 2024 Accepted: March 23, 2024 Online Published: April 04, 2024 doi:10.22158/eltls.v6n2p129 URL: http://dx.doi.org/10.22158/eltls.v6n2p129 Abstract The English way construction exhibits distinct constructional meaning and semantic features; however, the prepositional slot in this construction has not received sufficient attention, and there remains a lack of comparative research regarding different prepositions in this field. This study set out to examine the semantic classification, mapping features, cognitive factors in the subcategory English way into/to construction, and to improve the algorithm of ΔPratio. The research adopted a bottom-up corpus-driven approach based on the BNC and used HAC (Hierarchical Cluster Analysis) to explore the semantic categories of the verb collexemes and compare the lexical richness in these two constructions. The results revealed that at the level of linguistic structure, coercion and metaphor, as well as family resemblance, served as legitimization methods for verb entry into the construction, while the principles of economy and expressivity disclosed the intentions and motivations behind the semantic extensions of constructions at the pragmatic level. Additionally, this research found that the latter two principles were competitive in the interpretation process of way into/to construction, providing insights into understanding the psychological and social factors in language communication. Keywords English way construction, ΔPratio Algorithm, Pragmatic Motivation, Cognitive Grammar 1. Introduction After Salkoff’s (1988) observation on the usage of “way”, Jackendoff (1990) coined the term way construction and defined its prototypical meaning as a particular event construction, expressing “difficult spatial path displacement”. In terms of syntactic structure, it was represented as [SUBJi[V[POSSi-way]OBL]] (Goldberg, 1995). The construction consists of one verb and three www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 6, No. 2, 2024 Published by SCHOLINK INC. 130 arguments: [SUBJi], [POSSi-way], [OBL], representing the subject, possessive pronoun and an oblique prepositional phrase, respectively. Depending on the choice of different prepositions, English way into/to construction is considered subcategory of English way construction, as illustrated in example (1) and (2). When the verb itself lacks a sense of movement, especially when the argument does not indicate spatial location, as shown in example (1) and (2), the construction breaks through the inherent syntactic and semantic restrictions of the verb, forming the concept of atypical construction with a motion-meaning match. (1) A fake doctor tricked his way into a hospital job. (2) I made my way to this celebrated establishment. To date, the semantic classification of the English way construction has emerged as a focal point of research (Levin & Rappaport, 1988). Due to diverse viewpoints on classification, formulating a unified set of criteria became challenging. Consequently, incomplete existing classification systems fail to cover exceptional and newly emerging cases. Jackendoff (1990) and Goldberg (1995) both divided the semantic meaning of the English way construction into means and manner senses. However, it could be easily observed in the BNC and Google search that numerous counterexamples existed (Chen, 2010; Zhou, 2016). Diachronic research on the English way construction has revealed a third incidental sense apart from the means and manner senses in modern English, which referred to the incidental actions that occurred as the mover progresses along a path (Israel, 1996). Some domestic studies probed into the classes of verb collexemes based on Talmy’s (2000) cognitive semantic conceptualization lexicalization pattern theory (Chen, 2010; Gu & Lin, 2017). Although these studies have improved Goldberg’s semantic classification of verb collexemes to some extent (Zhang & Xu, 2013; Zhou, 2016; Zhang & Liu, 2017), relying solely on constructional collocation strength as the primary criterion for classification is somewhat simplistic. Another debate focused on the restrictive conditions for verbs entering the English way construction. Previous studies confined themselves to formal aspects of language, neglecting the impact of communicative contexts on its usage and thus impeding a comprehensive understanding of its pragmatic functions (Levin & Rappaport, 2005). From a generative grammar perspective, it held that only intransitive nonergative verbs other than nonaccusative verbs can be used in the English way construction. However, it faced multiple counterexamples. Mondorf (2010) was the first to illustrate the process of grammaticalization in the English way construction, in which the semantic meaning enhanced from concrete spatial entities to abstract or metaphorical non-spatial entities (Zhang & Xu, 2013). Following this line of thinking, subsequent research focused on the approach in expanding the semantic categorization network of the English way construction (Gu & Lin, 2017) and examining the cognitive rationale behind mapping abstract space and verb collexemes with non-movement meanings (Zhang, Yu, & Wu, 2019). Cognitive Construction Grammar emphasized that each construction should encompass syntax, semantics, and pragmatics, with pragmatic requirements serving as the fundamental driving force behind the evolution and syntactic variations of the English way construction. Although www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 6, No. 2, 2024 Published by SCHOLINK INC. 131 studies have employed principles such as metaphor and coercion to explain the validation of verb collexemes, limited investigation was conducted thus far into the pragmatic motivations. The aforementioned fruitful research payed attention to investigating the interaction between the verb and the English way construction, but overlooked the impact of different prepositions in the oblique slot. Moreover, there was a lack of comparative studies on the English way construction with different prepositions. Previous research examined the frequency and preference of verbs in the into construction (Hunston & Francis, 2000) and classified the semantics of causative predicates (Kim & Davies, 2015). When interpreting constructions semantically, attention should be paid to the interplay between the verb and other parts of the construction (Szczsniak, 2013). Many studies then started to observe the interaction between different slots in the construction (Gries & Stefanowitsch, 2003; 2004) and analyzed the co-occurring morphemes that vary together in the into construction (Yang & Wang, 2010; Zhang, Liu, & Liu, 2019). Besides, the research scope started expanding to multiple corpora (Rudanko, 2005; Kim & Lee, 2013). The abundance of research on the into construction indicated its significance and research value. As an integral component of the English way construction, its interaction with the English way construction should not be overlooked. In recent years, under the background of the quantitative and social turn in cognitive linguistics, construction grammar research has shown new development. Scholars such as Goldberg (1995) mainly centered on the verb slot in the English way construction while ignoring the interaction between the verb and the construction. Subsequent studies began to use Fisher’s exact test and employed the ΔPratio method to discuss the different dominant roles played by verb collexemes and the English way construction in directional collocations based on large-scale corpora (Wang & Lin, 2022). However, using this method may probably lead to the occurrence of many extreme p-values. After applying this method to select the access verbs based on p-values, the semantic classification relied on the researcher’s own understanding and judgment of the semantic characteristics of verbs, resulting in strong subjectivity in this introspection-based classification. Taking into account the above shortcomings, this research is thus motivated to contribute to this line of study by conducting a comparative analysis of atypical English way into construction and way to construction, treated as individual cases to explore their underlying pragmatic motivations. Improvements to the ΔPratio algorithm will help avoid extreme values in data analysis, and HAC (Hierarchical Clustering Analysis) can allow a more comprehensive examination of verb semantics with exceptions, thereby enhancing the flexibility and precision of the classification. This study not only provides new perspectives and improves methods in the field of linguistics but also contributes to a deeper understanding of the psychological and social factors involved in language communication. Research questions in this paper are as follows: 1) What are the frequency and semantic features of verb collexemes in the two subcategories of atypical English way construction? 2) What is the bidirectional interaction between the two subcategories of atypical English way www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 6, No. 2, 2024 Published by SCHOLINK INC. 132 construction and verb collexemes? 3) What are the pragmatic driving forces behind the two subcategories of atypical English way construction? 2. Data and Method The research consisted of three sequential steps. Firstly, an extensive search was conducted in the BNC (British National Corpus) to locate all occurrences of way into/to construction to identify the verb collexemes. Secondly, a categorization of the identified verb collexemes was carried out. Finally, a comprehensive analysis of the data was conducted. 2.1 Data In this investigation, BNC (British National Corpus), a collaborative effort between the University of Oxford and the University of Canterbury in the U.K. was applied. Access to the BNC online was available through its official website (https://www.english-corpora.org/bnc/), where the keywords “way into” and “way to” were typed in the search box. Then the obtained data were inputted into WordSmith Tools 6, followed by the application of the concordance tool in the software to identify co-occurring verb frequencies. The findings were organized by removing static verbs, such as “be”, “know”, “lose” and “have” (Yu, 2015). Subsequently, this study identified high-frequency verb collexemes, each with frequencies equal to or exceeding three. Our analysis yielded a total of 528 instances of way into construction and 629 instances of way to construction. To ensure the integrity and robustness of our results, this study also included complex linguistic structures, involving instances of “verb+one’s+other elements+way+into/to” as well as diversified morphological characteristics of the verb collexemes. For example, constructions such as “make its steady way into” and “buy not only her way into heaven but that of all her husband and relations” were carefully examined. Furthermore, in cases involving parallel verb collexemes, selection was placed on the verb most near to the word “way”. For instance, in the expression “grinning and spinning his way into”, only the verb “spin” was selected for further analysis. 2.2 Collostructional Analysis Method Stefanowitsch and Gries (2003) introduced a groundbreaking method known as Collostructional Analysis within the framework of Cognitive Construction Grammar. This method comprises three crucial components: Collexeme Analysis, Distinctive Collexeme Analysis, and Covarying Collexeme Analysis. It offers a more objective and rigorous approach to identifying the semantic characteristics of constructions by comparing the collocational strength between constructions and lexemes. Collexemes are words that exhibit significant associations with specific slots within a given construction. Collexeme Analysis aims to investigate the interplay between constructions and the words occupying those slots. This fundamental method for examining collocational patterns relies on a 2×2 contingency table, depicted in Table 1 (Gries & Stefanowitsch, 2010). To assess the statistical significance of the collocational strength between verb collexemes and www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 6, No. 2, 2024 Published by SCHOLINK INC. 133 constructions, we employed the collostructions package in the R 4.3.3 software. The p-value was calculated with the Fisher-Yates exact test. The collocational strength shows the bidirectional attraction between the verb collexemes and the English way construction. The higher the values, the stronger the association between the verb and the corresponding way construction is. Table 1. 2×2 Co-Occurrence Table for a Collexeme Analysis Word Target constructions Other constructions Total Words in slot 2 a b a+b Other words in slot 2 c d c+d Column Totals a+c b+d N=a+b+c+d 2.3 Improvements in ΔPratioMethods ΔP method can be used to assess the association between two constituents. Ellis and Ferreira-Junior (2009, p. 202) discovered that ΔP demonstrated a near-perfect ability to predict the initial choice of verbs in constructions by learners. Expanding on this notion, Gries (2013, pp. 143-144) advocated for using ΔP as an indicator of collocational strength, emphasizing its potential under the framework of corpus linguistics. Its directional nature and computational simplicity align with the psychological and psycholinguistic realities. The interaction between constructions and verbs not only generates a mutual attraction, but also gives rise to two distinct types of unidirectional collocational relationships (Wang, 2022). Based on the approach of constructional analysis, we further employ the ΔP method to analyze the strength of the unidirectional collocational relationship between the English way construction and verb collexemes. Using the phrase “due to” as an illustration, these two words demonstrate mutual collocability and directionality. The word “due” reliably co-occurs with “to”, forming the commonly encountered expression “due to”, whereas “to” exhibits a lower propensity to predict “due”. This exemplifies a fairly typical unidirectional collocational dynamic. Nevertheless, ΔP method, while proficient in gauging the direction and intensity of unidirectional associations, encounters difficulty when comparing the strengths of opposing unidirectional relationships in cases where one is not substantially stronger than the other. To meet this challenge, ΔPratio offers a solution (Wang & Lin, 2022). In ΔPratio, ΔPw→v shows the unidirectional attraction of the English way construction towards verb collexemes, whereas ΔP v→w suggests the reciprocal attraction of verb collexemes towards the English way construction. ΔP w→v = - and ΔP v→w = - . A positive sign indicates the dominance of the construction in the unidirectional relationship, whereas a negative sign suggests the predominance of the verb. The greater the absolute value of the ΔPratio, the more robust the unidirectional relationship’s attraction is. www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 6, No. 2, 2024 Published by SCHOLINK INC. 134 if ΔP w→v ≥ΔP v→w - if ΔP w→v ＜ ΔP v→w However, this formula suffers from an issue of exaggerated discrepancies. For instance, when comparing the ΔPratio values for “find” (151.5296066) and “barge” (0.99843015), we encounter a wide numerical range that produce extreme values. To meet this challenge, we made significant advancements to the original formula by incorporating the natural logarithm method. Specifically, our revised formula employs log(ΔPratio), where ΔPratio represents . This enhancement effectively mitigates the problem of extreme values in the ratio, allowing us to determine the collocational direction between verb collexemes and constructions. 2.4 Behavioral Profile Analysis Due to the inherent flexibility and contextual variations in the usage of verb collexemes in the English way construction, this study employed the BP (Behavioral Profile) analysis. The BP analysis combines manually annotated features with multivariate statistical techniques, presenting a corpus-based approach that mitigates biases arising from intuition and frequency of use (Divjak & Gries, 2009; Gries & Divjak, 2009). This methodology comprises two primary components: ID labels and behavioral characteristics. ID labels (Gries, 2006; Gries & Divjak, 2009) include diverse linguistic features like formal, semantic, and pragmatic aspects, which are identified through manual annotation of instances within a corpus. The BP analysis involves several sequential steps. Firstly, instances were collected to create datasets that reflected the authentic usage of verb collexemes in two subcategory way constructions. Subsequently, the gathered sentences were manually annotated based on the construction’s position, the tense of the sentence, the form of the verb collexemes, the presence or absence of an animate subject, and features of the subject mapping domain. Using the annotated data, HAC (Hierarchical Agglomerative Clustering) was employed to reveal semantic relations by visualizing hierarchical clustering. HAC served as an exploratory technique, illustrating the similarity between given forms through the identification of hierarchical clustering relationships. The resulting analysis was presented with a dendrogram. 3. Results and Discussion 3.1 Collocation of verb Collexemes in the Subcategory English Way Construction As depicted in Table 2, log(ΔPratio) values for “wheedle”, “barge”, “bluff”, “grope” and “fumble” are all negative, indicating that the construction has less attraction from the verb collexemes compared to the verbs that attract the construction. Specifically, when considering the verb “make” in the way into construction, the ΔPw→v value is 0.068550771, the ΔPv→w value is 0.0002086, and the log(ΔPratio) is www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 6, No. 2, 2024 Published by SCHOLINK INC. 135 2.516698713. These results suggest that the construction demonstrates a significantly stronger inclination towards the verb “make”. Therefore, the construction exhibits substantial predictive power in the directional relationship with the verb “make”. It is noteworthy that despite the relatively low p-value of 17.31270 for the verb “feel”, log(ΔPratio) remains remarkably high at 2.018004508. This finding underscores the predictive power of way into construction for the verb “feel”, emphasizing its dominant role in the directional relationship. In Table 2, p-values for “con” and “wheedle” are closely aligned at 116.71664 and 102.22216, while the corresponding log(ΔPratio) values are 0.183078347 and -0.864617791. This indicates that “con” is more predisposed to enter and form the way construction, while “wheedle” shows greater attraction to this construction, thus making it more likely for the construction to become a typical usage of “wheedle”. Table 2. The Calculation of Collocation Strength in the EnglishWay into Construction Verb P-value ΔPw→v ΔPv→w Log(ΔPratio) 1 Find 2714.47029 0.388362199 0.002562946 2.180497482 2 Force 520.99947 0.085986806 0.00153869 1.747280683 3 Worm 312.30307 0.034524052 0.016839179 0.311800839 4 Make 249.38957 0.068550771 0.0002086 2.516698713 5 Push 244.26610 0.03757101 0.00226123 1.220508229 6 Con 116.71664 0.014119109 0.009262515 0.183078347 7 Wheedle 102.22216 0.009418342 0.068959207 -0.864617791 8 Claw 88.04101 0.010979921 0.007635621 0.15775483 9 Edge 85.66430 0.015601097 0.00101832 1.185270767 10 Bluff 83.82977 0.009415292 0.015299812 -0.210852301 11 Smash 76.85475 0.010968741 0.003435195 0.504205449 12 Hack 76.56224 0.009412052 0.008373578 0.050773247 13 Fumble 65.66795 0.007844394 0.01009469 -0.109533617 14 Trick 61.42593 0.009393973 0.002370869 0.597941723 15 Shoulder 56.92226 0.010904328 0.000819561 1.124017473 16 Weave 56.29411 0.007836734 0.003958787 0.296572944 17 Grope 54.01718 0.006276185 0.012151725 -0.286942197 18 Buy 49.47135 0.012299577 0.000302066 1.609787611 19 Cheat 44.14088 0.006268205 0.003542916 0.247782356 20 Fight 42.19068 0.009292858 0.000468486 1.297452441 21 Work 38.61414 0.015838344 7.05417E-05 2.35126362 22 Barge 34.80789 0.004703246 0.004710641 -0.000682312 23 Ease 33.52931 0.006236897 0.000933078 0.825050499 www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 6, No. 2, 2024 Published by SCHOLINK INC. 136 24 Kick 31.93047 0.006227418 0.000762014 0.912344889 25 Inch 23.53587 0.004667808 0.000711363 0.817021463 26 Talk 22.30557 0.007457557 0.000121251 1.788912738 27 Feel 17.31270 0.007185385 6.89359E-05 2.018004508 28 Win 12.56807 0.004439089 0.000104529 1.628057577 As shown in Table 2 and 3, the subcategories of the English way construction share 11 high-frequency verb collexemes. The p-values for “make”, “grope” and “fight” in the English way to construction markedly surpass their counterparts in the English way into construction, indicating their stronger association with way to construction. Regarding the prepositional slot, our analysis reveals that in the way to construction, the frequency of “make” (55.4%) overwhelmingly prevails “find”, corroborating the findings of Zhang and Xu (2013). However, in the way into construction, we observe a notably higher frequency of “find” compared to “make”, the former being 5.5 times more prevalent. “Find” indicates a discovery, yet unlike “make”, it doesn’t mean creating something from nothing; rather, it unveils something existing but previously undiscovered, both of which can be considered forms of creation. It is notable that from a diachronic perspective, the word “make” has been consistently used in the English way construction since the 15th century, surpassing all other verbs in frequency by the 16th century. Sullivan’s (2007) corpus analysis spanning the 11th to the 19th century corroborated the enduring prevalence of “make” in the English way construction. Nonetheless, over time, this usage has gradually waned, supplanted by verbs conveying the notion of progression along a path. Hence, it is unsurprising that the primacy of “make” in Table 2 has relinquished its top position. Table 3. The Calculation of Collocation Strength in the EnglishWay to Construction Verb P-value Δ Pw→v Δ Pv→w Log(Δ Pratio) 1 Make 3437.03576 0.552755648 0.001658156 2.522907893 2 Find 1231.1886 0.215253454 0.001404323 2.185482987 3 Grope 184.53171 0.019074611 0.035184446 -0.265895016 4 Fight 181.75138 0.031670148 0.001574063 1.303627901 5 Push 143.83292 0.025331363 0.001503162 1.226652884 6 Wing 117.00611 0.019023864 0.002209487 0.935007221 7 Claw 71.43034 0.009529791 0.006501362 0.166079003 8 Wend 65.62107 0.00635864 0.057136607 -0.953550178 9 Elbow 52.08505 0.007933086 0.00310128 0.407901143 10 Wind 51.16579 0.00635538 0.01009476 -0.200954448 11 Sing 44.41572 0.009451488 0.000679249 1.143471003 www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 6, No. 2, 2024 Published by SCHOLINK INC. 137 12 Bomb 40.04546 0.007895468 0.000924685 0.931383913 13 Fumble 35.23375 0.004764525 0.006017836 -0.101420734 14 Work 34.91860 0.016057865 7.06158E-05 2.356785677 15 Dig 34.3829 0.006327361 0.001244532 0.706216657 16 Battle 27.21635 0.006280527 0.000501235 1.097954287 17 Blast 27.09567 0.004750246 0.001551372 0.486000005 18 Power 26.97099 0.00915307 0.000149174 1.787872995 19 Bar 24.96387 0.006254491 0.000375211 1.221916402 20 Feel 20.80367 0.008874992 8.40689E-05 2.023532453 21 Dance 18.58434 0.004689062 0.000366642 1.106843197 22 Win 17.58431 0.006088784 0.000141553 1.633610507 23 Edge 17.44407 0.004671885 0.000301023 1.190892977 Owing to their fixed and conventionalized characteristics, prototypical verbs effortlessly establish collocational relationships with the English way construction. Verbs such as “find”, “make”, “force”, “push” and “worm” commonly exist in the English way construction, displaying prominent fixed and accepted attributes. As a result, these verb meanings easily blend with the construction’s meaning, emphasizing the concept of a “path” and facilitating a prototypical collocational association with the English way construction. However, it is worth noting that verbs in the dominant position of a bidirectional collocation relationship generally exhibit lower collocation strength and typically serve as non-prototypical verbs within the construction. Metaphor and coercion serve as techniques for verbs to be accepted in constructions, enhancing the productivity of the English way construction. Even though the verbs and their conceptual domains in atypical sentences may lack spatial motion relationships, they can establish abstract collocations through noun metaphors or verb metaphors (Israel, 1996). Verbs such as “con”, “bluff”, “trick”, “wheedle” and “cheat” are inherently associated with the meanings of “deception” and “intimidation”, posing challenges to their compatibility with the notion of a path in the way into construction. Only when these verbs enter the construction and undergo semantic suppression can they acquire the corresponding sense of motion through metaphorical interpretation, rendering the entire sentence coherent and appropriate (Zhou &Wang, 2017; Lin & Wang, 2013; Lin & Zhang, 2020). 3.2 The Distribution and Semantic Characteristics of Verb Collexemes in the English Way Construction HAC was applied to investigate the distribution and semantic characteristics of high-frequency verb collexemes in the English way into/to construction. The semantic cluster analysis for the verb collexemes in the way into construction are shown in Figure 1, while Figure 2 illustrates the findings for the way to construction. These verbs in two constructions are classified into six clusters. It is noteworthy that the way into construction demonstrates a broader and more prolific spectrum of verb collexemes compared to the way to construction. Notably, within both clustering diagrams, “work” and www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 6, No. 2, 2024 Published by SCHOLINK INC. 138 “feel”, as well as “win” and “edge” are grouped together in the same cluster, suggesting analogous patterns of usage across both subcategories of the English way construction. Figure 1. Semantic Clustering Graph of High-Frequency Verb Collexemes in theWay into Construction Based on the data presented in Table 4, the type-token ratio in the sixth category of the way into construction accounts for 28.6% of the total, with the second, third, and fifth categories following closely and the fourth category has the lowest proportion, comprising only 1.8%. The distribution of the type-token ratio in the way to construction has a similar pattern. Notably, when considering the 11 shared words between both constructions, the way into construction demonstrates a significantly higher usage of inanimate subjects compared to the way to construction. This disparity is particularly notable for verbs like “make”, “find”, “fight”, “push” and “fumble”, yielding a substantial increase in the number of mappings to the abstract domain in the way into construction compared with the way to construction. This observation suggests an expansion of the mapping domain in the way into construction, extending from physical space to abstract space. www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 6, No. 2, 2024 Published by SCHOLINK INC. 139 Figure 2. Semantic Clustering Graph of High-Frequency Verb Collexemes in theWay to Construction Some verbs of the second category in the way into construction and the third category in way to construction associate with coercive control, such as “hack”, “force”, “bar” and “battle”. Additionally, the second category of way into construction also includes manipulative words like “bluff”, “wheedle”, “con” and “trick”, but there are no equivalents in the way to construction. Both coercion and manipulation are means to achieve specific aims. While coercion typically involves direct and forceful actions, manipulation tends to use psychological and informational tactics to influence others’ cognition or behavior, which reflects a nuanced selection of control methods. Consequently, the verbs in the way into construction exhibit a tendency towards abstraction and expansion along psychological pathways. Table 4. Clustering Categories of the Verb Collexemes in two Subcategories ofWay Construction The Englsih way into construction Type-token ratio The English way to construction Type-token ratio 1 Inch (3), worm (22),claw (7),find (248),win (3) 1.8% Fumble (3), elbow (50), grope (12), make (349) 1.4% 2 Bluff (3), ease (4), hack (6), wheedle (6),make (45), force (55), push (24), con (9), trick (6) 5.7% Feel(6),power(6),find(136),fight(20),work( 11) 2.8% 3 Grope(4),feel(5),shoulder(7),talk(5),wea ve(5),work(11) 16.2% Bar(4),push(16),battle(4),dig(4) 14.3% www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 6, No. 2, 2024 Published by SCHOLINK INC. 140 4 Fumble(5),edge(10),smash(7) 13.6% Dance(3),wing(12) 13.3% 5 Cheat(4),buy(8),fight(6) 16.7% Wend(4),blast(3),sing(6),bomb(5),wind(4) 22.7% 6 Barge(3),kick(4) 28.6% Claw(6),edge(3),win(4) 23.1% There are a total of 79 instances in the way to construction that enter the non-physical domain, accounting for 12.5% of the total. In contrast, out of the 349 instances of the verb “make”, only 22 cases enter the non-physical domain. While in the way into construction, there are 193 instances where “into” is followed by a non-physical domain, accounting for 36.5% of the total. Among the words on the right side of “into”, the words related to the mental domain are the most abundant (e.g., “heart” 8 instances, “consciousness” 6 instances, “affection” 6 instances, “confidence” 7 instances, “mind” 4 instances), followed by the economic domain (e.g., “job” 8 instances, “business” 4 instances, “account” 4 instances, “employment” 1 instance) and the legal domain (e.g., “final” 7 instances, “system” 7 instances, “side” 6 instances, “position” 5 instances). The investigation found that in the way into construction, the verb “find” and “worm” had a significantly higher number of instances enters the non-physical domain, with 110 and 20 instances respectively, much higher than the other entry verbs. Especially, the word “worm” forms idiomatic expressions such as “worm one’s way into one’s heart” (6 instances) and “worm one’s way into one’s confidence” (4 instances), thus enhancing the productivity of the way into construction. When assessing the validity of a verb’s integration into the construction, two primary considerations may come into play: the application of metaphor and coercion. The degree of similarity between new words and previous terms is also a criterion for speakers to judge the entry of verbs into construction (Suttle & Goldberg, 2011). Furthermore, recognizing an integrated knowledge network involves continuously refining and expanding existing cognitive frameworks by reinforcing or introducing new connections and conceptual representations, while also adjusting previously established associations in response to new experiences (Goldberg, 2019). www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 6, No. 2, 2024 Published by SCHOLINK INC. 141 Figure 3. The Semantic Network of the Second Category of the verb Collexemes in the Way into Construction The productivity of the way construction is possibly triggered by two main factors. Firstly, certain verbs attract others of the same semantic feature into the construction. For instance, verbs like “shoulder” and “elbow” have p-values of 56.92226 and 52.08505 respectively, with log(ΔPratio) values of 1.124017473 and 0.407901143. Since “shoulder” belongs to the category of body parts, another body part “elbow” becomes easy to enter the construction. Secondly, the mapping domain of a construction forms a semantic network. Larger networks having more interconnection and associations. For a verb to enter the construction, it depends on the degree of overlap between its mapping domain and those already existing in the construction. The greater the overlap, the easier it is for the verb to enter the construction. As shown in figure 3, when speakers assess whether a verb can enter the construction, they compare the similarity between the characteristics of its mapping domain and a prototype. This similarity can include all members of a category or just some members. In the latter case, category consistency arises from family resemblance. The typical mapping domain of the way construction consists of concrete spatial entities such as “home, “house” and “room”, with abstract or metaphorical non-spatial entities like “job”, “life” and “favor” extending from this foundation as atypical mapping domains. 3.3 Pragmatic Motivation of Atypical English Way Construction Conventional expressions often fall short in accurately and vividly conveying the speaker’s intended meaning. To address this problem, individuals continually introduce constructions like the English way construction, which provides a concise and efficient means of expressing emotions while underscoring the significance of conveying precise meaning. From a pragmatic perspective, people strive to use more concise, innovative or universal linguistic units to meet the needs of communication and expression. It can be imagined that without constructions www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 6, No. 2, 2024 Published by SCHOLINK INC. 142 like example (3) and example (5), people would need to use sentences like example (4) and example (6), which contain more words to express similar meanings. Both example (3) and example (4) describe the achievement of economic success or wealth through increased spending. “Spending our way” implies the initiative and effort exerted by the subject, emphasizing that individuals or governments need to take proactive action to promote economic growth and prosperity. In contrast, “achieve economic success or wealth by increasing spending” only states the goal of achieving economic success or wealth through increased expenditure, without emphasizing the proactive nature or effort involved. At the same time, although both example (5) and example (6) can express a slow and effortless manner, the word “ease” better conveys the leisurely and relaxed mood of the elderly lady, adding a more delightful expressive effect. (3) The big government theory says we can hamstring business and tax and spend our way to prosperity. (4) The big government theory says we can hamstring business and tax and achieve economic success or wealth by increasing spending. (5) An elderly lady eased her way into the compartment. (6) An elderly lady walked slowly, less painfully into the compartment with less effort. The economy principle, proposed by Leech (1983) and supported by Langacker (1999; 2008), suggested that the optimal grammar employs the fewest symbols necessary to achieve its communicative objectives. Streamlining text without losing information reduces the cognitive load in both encoding and decoding processes. Nevertheless, an overly pursuit of economy may render the text cryptic and challenging to comprehend. Hence, while striving for temporal and cognitive efficiency, a balance must be struck between brevity and clarity. Leech (1983) further advanced the expressivity principle, advocating for the richness and adaptability of linguistic expression. Language not only conveys factual information but also the speaker’s emotions, attitudes, and personality, going beyond mere utility in saving time and energy. The English way into construction and way to construction simultaneously encompass the principles of economy and expressivity. In the two subcategories of way construction, although both principles are considered, whether they are competitive or not needs further verification. In this paper, accordingly, expressivity is operationalized as the sentiment score from Sentiment R, ranging from -1 to +1. Negative values indicate negative sentiment, positive values indicate positive sentiment, and values close to zero indicate neutral sentiment. The operational definition of the economy principle is the average sentence length, with shorter sentence lengths reflecting a more pronounced economy principle. Additionally, higher absolute values of sentiment scores reflect a more pronounced expressivity principle. As outlined in Table 5, a detailed comparison of 11 shared verb collexemes in both way into construction and way to construction unveils that verbs such as “make”, “find”, “claw”, “push”, “work”, “win”, “fight”, “grope” and “fumble” have higher sentiment scores in the way into www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 6, No. 2, 2024 Published by SCHOLINK INC. 143 construction. Based on annotated factors, this investigation demonstrates that verb collexemes with a greater prevalence of abstract mapping domains within distinct subcategories of the English way construction show elevated sentiment scores, underscoring the dominance of the expressivity principle in linguistic competition. For instance, the verb “find”, when utilized in the way into construction, allocates 44.3% of its usage to abstract domains, in contrast to the mere 8.1% observed in the way to construction. Likewise, the verb “make” exhibits a distribution of 24.4% towards abstract domains within the English way into construction, compared to a modest 6.3% in the way to construction. Conversely, the sentiment scores for “feel” and “edge” manifest an opposite trend. In the way into construction, “feel” attributes 40% of its usage to abstract domains, while this percentage rises to 60% in the way to construction. Similarly, “edge” accounts for a mere 10% in the way into construction, yet comprises 100% in the way to construction. This observation suggests that the dominance of the expressivity principle, characterized by a higher prevalence of abstract mapping domains, is markedly evident across two subcategories of the English way construction when using the same verb. Table 5. The Average Sentence Length and the Average Absolute Value of Sentiment Scores for the 11 Shared Verb Collexemes in Two Subcategory Constructions Way into Average sentence length Sentiment score Way to Average sentence length Sentiment score Make 17.04444444 0.126297055 Make 18.32276657 0.114017961 Find 17.65725806 0.238655621 Find 17.79411765 0.153328287 Claw 14.71428571 0.230895941 Claw 16.66666667 0.135035261 Push 16.50000000 0.140305302 Push 17.12500000 0.135412583 Work 15.18181818 0.247997764 Work 18.54545455 0.243387568 Feel 15.60000000 0.12909944 Feel 14.66666667 0.201795839 Win 17.66666667 0.236558413 Win 18.67833383 0.177215432 Fight 20.33333333 0.278138105 Fight 20.95000000 0.194049491 Grope 15.25000000 0.284471516 Grope 16.91666667 0.146041269 Fumble 18.75000000 0.21891145 Fumble 19.00000000 0.181405004 Edge 14.20000000 0.074503436 Edge 17.33333333 0.152883752 Moreover, upon closer examination within the same construction, it is evident that, with the exception of “edge” in the English way into construction, the remaining verbs demonstrate higher sentiment scores than “make”. This difference implies a greater intensity in the expression of emotions, consequently enhancing perceptibility and emotional resonance in communication. A fundamental goal of human discourse and expression revolves around conveying emotions and portraying sensory experiences related to the world. By using words imbued with strong emotions, a vividness in www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 6, No. 2, 2024 Published by SCHOLINK INC. 144 emotional expression can be achieved. 3. Conclusion This study, drawing upon the BNC, employed a cognitive construction grammar approach to identify verb collexemes and examine their unidirectional interaction in two distinct subcategories of the English way construction. By expanding the analysis scope to encompass the prepositional slot, it found cognitive and pragmatic driving forces underlying complex and atypical English way into/to construction instances. Furthermore, the study enhanced the ΔPratio method originally proposed by Wang and Lin (2022) and applied HAC analysis to grope verb collexemes into semantic categories. By studying the collocation strength and directional tendencies between core verbs and way construction, this investigation revealed that verb collexemes in way into construction typically demonstrated higher productivity compared to their counterparts in the English way to construction. Moreover, the higher level of abstraction, the conspicuous influence of the expressivity principle becomes. People chose emotional words to express ideas in a more vivid manner, which serves as one of the pragmatic motivations. Subsequent research on collocational constructions could make more efforts in considering contextual nuances, target audience specifications, and cultural backgrounds, recognizing that sentiment analysis alone does not capture the entirety of emotional expression. Acknowledgement My deepest gratitude goes first and foremost to my respectable supervisor, Huang Ying, for his generous guidance, expertise and continuous support throughout this journey from the selection of the topic to the completion of the study, as well as his patient guidance and selfless help in teaching me the statistical software. His insightful feedback and encouragement have greatly shaped the direction of my research and his diverse perspectives and knowledge have enriched my work and broadened my understanding of the subject matter. My sincere thanks also go to all the other professors and teachers in the School of Foreign Studies, CUMT, who have taught and helped me in my postgraduate study. Their pioneering research and publications have been instrumental in shaping my research and expanding my horizons. In conclusion, I would like to once again express my sincere appreciation to everyone who has contributed to my academic journey. Their support and assistance are crucial in making this research possible. References Divjak, D., & Gries, S. T. (2009). Corpus-based cognitive semantics: A contrastive study of phasal verbs in English and Russian. In Dziwirek, K., & Lewandowska-Tomaszczyk, B. (Eds.), Studies in Cognitive Corpus Linguistics (pp. 273-296). Peter Lang, Lausanne. Ellis, N. C., & Ferreira-Junior, F. (2009). Constructions and their acquisition: Islands and the www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 6, No. 2, 2024 Published by SCHOLINK INC. 145 distinctiveness of their occupancy. Annual Review of Cognitive Linguistics, (7), 187-220. https://doi.org/10.1075/arcl.7.08ell Goldberg, A. E. (1995). Constructions: A Construction Grammar Approach to Argument Structure. Chicago: University of Chicago Press. Goldberg, A. E. (2019). In Explain Me This: Creativity, Competition and the Partial Productivity of Constructions (pp. 61-70). Princeton: Princeton University Press. https://doi.org/10.2307/j.ctvc772nn Gries, S. T. (2006). Corpus-based methods and cognitive semantics: The many senses of to run. In Gries, S. T., Stefanowitsch, A. (Eds.), Corpora in Cognitive Linguistics: Corpus-Based Approaches to Syntax and Lexis (pp. 57-99). De Gruyter Muton, Berlin. https://doi.org/10.1515/9783110197709.57 Gries, S. T. (2013). 50-something years of work on collocations: What is or should be next ….International Journal of Corpus Linguistics, 18(1). 137-165. https://doi.org/10.1075/ijcl.18.1.09gri Gries, S. T., & Divjak, D. (2009). Behavioral profiles: A corpus-based approach to cognitive semantic analysis. In: Evans, V., Pourcel, S. (Eds.), New Directions in Cognitive Linguistics (pp. 57-75). John Benjamins, Amsterdam. https://doi.org/10.1075/hcp.24.07gri Gries, S. T., & Stefanowitsch, A. (2004). Covary collexemes in the into-causative. In M. Achard, & S. Kemmer (Eds.), Language,Culture, and Mind［C］ (pp. 225-236). Stanford CA: CSLI. Gries, S. T., & Stefanowitsch, A. (2010). Cluster analysis and the identification of collexeme classes. Empirical and Experimental Methods in Cognitive/Functional Research (pp. 73-90). Gries, S. H. (2013). This volume. More (old and new) misunderstandings of collostructional analysis: On Schmid & Küchenhoff. Hunston, S., & Francis, G. (2000). Pattern Grammar: A Corpus-driven Approach to the Lexical Grammar of English. Amsterdam and Philadelphia: John Benjamins Publishing Company. https://doi.org/10.1075/scl.4 Israel, M. (1996). The English way-constructions Grow. In A. Goldberg (Ed.), Conceptual Structure, Discourse and Language[C] (pp. 217-230). Stanford: Center for the Study of Language and Information Publications. Jackendoff, R. S. (1990). Semantic Structures. Cambridge, MA. : The MIT Press. Kim, J. B., & Davies, M. A. (2015). The into-causative construction in English: A construction-based perspective . English Language and Linguistics, (1), 1-29. https://doi.org/10.1017/S1360674315000271 Kim, J. B., & Lee, N. G. (2013). The transitive into-ing construction in English: A usage-based approach. English Language and Linguistics, (13), 395-418. https://doi.org/10.15738/kjell.13.2.201306.395 Langacker, R. W. (1999). Grammar and Conceptualization. Berlin: Monton de Cruyter. www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 6, No. 2, 2024 Published by SCHOLINK INC. 146 https://doi.org/10.1515/9783110800524 Langacker, R. W. (2008). Cognitive Grammar: A Basic Introduction. New York: Oxford University Press. https://doi.org/10.1093/acprof:oso/9780195331967.001.0001 Leech, G. (1983). Principles of pragmatics. Longman, 64-70. Levin, B., & Rappaport, M. (2005). Argument Realization. New York: Cambridge University Press. https://doi.org/10.1017/CBO9780511610479 Levin, B., & Rappaport, T. (1988). Lexical subordination //In BRENTARI D,LARSON G,MACLEOD L(eds). Proceedings of the 24th Annual Meeting of the Chicago Linguistic Society. Chicago: Chicago linguistic society, 1988: 275-289. Lin, Z., & Wang, K. (2013). On the rationality of atypical complex constructions. Modern Foreign Languages, 36(04), 363-370+438. Lin, Z., & Zhang, H. (2020). Modeling the semantic interaction of word collocation constructions: A case study of "Adj. + N." Foreign Languages, (6), 64-72. Mondorf, B. (2010). Variation and change in English resultativeconstructions. Language variation and change, 22(3), 397-421. https://doi.org/10.1017/S0954394510000165 Rudanko, J. (2005) Lexico-grammatical innovation in current British and American English: A case study on the transitive into-ing pattern with evidence from the bank of English Corpus. Studia Neophilologica, (77), 171-187. https://doi.org/10.1080/00393270500384791 Salkoff, M. (1998). “Analysis by Fusion”. Lingvisticae Investigation, (1), 49-84. https://doi.org/10.1075/li.12.1.03sal Stefanowitsch, A., & Gries, S. T. (2003). Collostructions: Investigating the interaction of words and constructions. International Journal of Corpus Linguistics, 209-243. https://doi.org/10.1075/ijcl.8.2.03ste Sullivan, K. (2007). Grammar in metaphor (PhD. Dissertation). Berkeley: University of California. Suttle, L., & Goldberg. A. E. (2011). “The Partial Productivity of Constructions as Induction.” Linguistics, 49(6), 1237- 69. https://doi.org/10.1515/ling.2011.035 Talmy, L. (2000). Toward a Cognitive Semantics (Vol 1), [M]. Cambridge: MIT Press. https://doi.org/10.7551/mitpress/6847.001.0001 Wang, H., & Lin, Z. (2022). A study on the collocational relationship between verbs and directional constructions in English “Way” constructions. Journal of PLA University of Foreign Languages, 45(04), 26-34+87+160. Yang, J., & Wang, Y. (2010). Co-occurring morphemes in into-causative constructions: A review of foreign research on into-causative constructions. Foreign Language Studies, (02), 70-75. Yu, C. (2015). Research on the relexicalization pattern of way-construction in Chinese based on English-Chinese parallel corpus. Foreign Language Teaching and Research, 42-54, 159-160. Zhang, J., & Liu, L. (2017). A study on the linguistic features of way-construction. Journal of Shanghai University of International Business and Economics, 24(06), 75-84. www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 6, No. 2, 2024 Published by SCHOLINK INC. 147 Zhang, J., & Xu, Y. (2013). Discussion on the grammaticalization process of WAY constructions. Foreign Language Teaching, 34(06), 1-6[4]. Zhang, J., Liu, L., & Liu, P. (2019). A study on the linguistic features of causative constructions based on the contemporary American English corpus: Taking into-V-ing causative constructions as an example. Foreign Language Teaching, 40(04), 34-39. Zhang, Y., Yu, C., & Wu, X. (2019). Cognitive study on the mismatched motion semantics of 'way' constructions based on COCA corpus. Journal of PLA University of Foreign Languages, 42(04), 85-93+160. Zhou, X. (2016). Reconsideration of the verb admission mechanism in Way-construction: Theoretical explanation based on Levin and Rappaport. Foreign Languages and Teaching, (03), 28-36+145. Zhou, X., & Wang, J. (2017). Two-dimensional analysis of the verb suppression of 'way' construction bodies. Foreign Languages and Their Teaching, (5), 64-72. Yuting Gu1* & Ying Huang2 1 School of Foreign Studies, China University of M Abstract References