Articles
Correspondence Analysis of Tennyson’s Characteristic Use of Modal Adverbs
2025 年 7 巻 1 号 p. 1-10
詳細
2025 年 7 巻 1 号 p. 1-10
This research aims to investigate the characteristic use of modal adverbs in Alfred Tennyson’s poetry. Numerous studies on Tennyson’s poems have paid critical attention to stylistic similarities between Tennyson and other poets; however, previous studies have seldom referenced Tennyson’s lexical choices by their parts of speech. Previous studies allude to, examine, or critique Tennyson’s style, syntax, or language, and they frequently concentrate on individual words that appear in certain poems, e.g., sea, nothing, or sun. Using a quantitative approach called correspondence analysis (CA), this study examines Tennyson’s distinctive modal adverbs and discovers that, when compared to the poems of eleven other Augustan, Romantic, and Victorian poets, maybe and mayhap are the most common modal adverbs in Tennyson. The discussion examines why the words maybe and mayhap are prevalent in Tennyson’s poetry. Although this study focuses only on fifteen modal adverbs and eleven works of referenced poets, it nonetheless discloses several novel characteristics of Tennyson’s style and demonstrates how effective correspondence analysis is as a tool in investigating verse texts as well as for reading into “unread” portions of literary works.
Alfred, Lord Tennyson (1809–1892), was a remarkable English poet and the British poet laureate (1850–1892). Since the 1800s, his poetry has been criticized and studied continuously. Studies on Tennyson’s poems and writing style abound (Shaw 1976; Hair 1991; and Thomas 2019, to name but a few). Many studies, particularly those focusing on his style, have discussed Tennyson’s stylistic resemblance to other authors.
Research on large amounts of text data from various corpora has been facilitated by developments in computer technology. In the fields of linguistics and literature, statistical methods have been applied to research, and numerous stylistic studies that utilized these quantitative, stylometric approaches have been published. Statistical methods enable the exploration of new dimensions: for example, results revealed using big data or “reading” (analyzing) multiple documents at once, that are difficult to identify with conventional research in the literature. Moretti (2013, 45) notes that a significant portion of primary source materials remains unexamined by scholars engaged in conventional literary studies, referring to the term “great unread”, coined by Margaret Cohen. Employing a quantitative technique in the analysis of poems, I attempt to reveal new insights into Tennyson’s works and highlight what has been left as the “great unread” by qualitative approaches (Moretti 2013, 45). However, few studies have employed such quantitative methods to study Tennyson’s writing. In this paper I apply a statistical analysis approach, correspondence analysis (CA), to investigate modal adverbs (ModAdv) such as perhaps, maybe, and haply. To shed light on Tennyson’s characteristic uses of ModAdvs, the work of other poets from the eighteenth and nineteenth centuries are considered. The results of CA suggest that two ModAdvs in Tennyson’s poetry, maybe and mayhap, stand out from other poets’ usages; thus, we can consider maybe and mayhap to be characteristic of Tennyson’s writing style. In addition to the results in terms of ModAdvs, this paper explains the effective use of statistical analysis on poems.
Numerous previous studies have investigated Tennyson’s poems, style, and even his life. Numerous studies have discussed specific words, themes, or concepts in Tennyson’s poems, e.g., Nakamura (1967), Hair (1991), Kabata (2001, 2007), Noguchi (2011), and Thomas (2019). Nakamura (1967) discusses Tennyson’s Christianity, and Hair (1991) examines vocabulary employed in Tennyson’s poetry to reveal his linguistic viewpoint. Hair highlights Tennyson’s connection to linguistic publications from his time and to other poets, including Homer, Samuel Taylor Coleridge, and Robert Browning. Through his analysis, Hair illustrates Tennyson’s adept use of the divine or “I” perspective in his works to convey his perspective of language. Kabata (2007) points out that Tennyson uses sun imagery to symbolize King Arthur’s prosperous reign and reinforce feelings of loneliness in unhappy women in “Mariana” (1830), “The Lady of Shallot” (1833), and “The May Queen” (1833). Additionally, according to Noguchi (2011), Ulysses’ journey in “Ulysses” (1842) is not merely a pursuit of knowledge, but also reflects Tennyson’s desire for life. Thomas (2019) centers her research on illustrating Tennyson’s echoes to William Wordsworth, who was Tennyson’s predecessor as the British poet laureate.
Shaw (1976) investigates similarities between Tennyson and other historical writers. Specifically, Shaw identifies eighteenth-century poets such as John Keats, Alexander Pope, and Percy Bysshe Shelley as having influenced Tennyson’s early poems. Shaw further suggests that the Victorian authors, including Tennyson, attempt “to keep the revolutionary Romantic and the conservative eighteenth-century elements in balance” (Shaw 1976, 21–22).
As discussed by Leech (1969, 13), not only Tennyson and other Victorian poets but also poets from various historical periods have exhibited a penchant for, or a readiness to adopt, the archaic mannerisms of their predecessors in poetry. Leech further suggests that this archaism manifested in both stylistic choices and lexical usage. Therefore, this paper posits the existence of stylistic similarities among these poets, while acknowledging the undeniable presence of differences at the lexical level. These distinctions can serve as valuable elements for effectively characterizing the works of each poet.
While various themes, concepts, and words used in Tennyson’s work have been studied extensively, his use of adverbs has received little attention. This paper specifically examines ModAdvs, a type of adverb that expresses the mental attitude and epistemic possibility of the narrators and characters in literary works.
This section briefly contemplates how ModAdvs are named or classified by grammarians to indicate how they can be identified in the context of this study. Sweet (1900, 119) does not use the term modal, but rather states, “Adverbs of assertion express affirmation, such as yes, yea; denial or negation, such as no, nay, not; asseveration, including certainty, doubt, etc., such as surely, certainly, assuredly, truly, undoubtedly, indeed, perhaps, possibly.” Swan (1995) uses the term “adverbs of certainty” and gives corresponding examples, including certainly, definitely, clearly, obviously, and probably. Kruisinga (1932, 124) categorizes three kinds of modality and described them as “adverbs of modality.” Greenbaum (1969, 94–98) and Quirk et al. (1985, 615) refer to such adverbs as “disjunct,” and Zandvoort (1975, 249) uses the term “Adverbs of modality.” “Modal adverb” began to be used in the 1990s, for instance, by Declerck (1991), Biber et al. (1999), and Huddleston and Pullum (2002). After reviewing these previous studies, I chose fifteen ModAdvs for further investigation.
With ongoing developments in computer technology, the paradigm of literary and linguistic research has undergone various changes. Statistical methods and their application in literary studies facilitate the evaluation of a huge number of works at once using quantitative methodologies. Qualitative research, or close reading, which refers to scholars attentively reading an author’s work, has been the primary approach to studying literature. The discussions and conclusions drawn from attentive reading are crucial and important in literature research. However, according to Moretti (2013, 48), the issue with close reading is that it inevitably depends on a very small canon. He further indicates that close reading leaves a “great unread”1 of the material in question because humans can only consider a small volume of work at a time. Note that the goals of Moretti (2013) and this study are not to critique close reading–based research nor the methodology in general. The purpose of this study is to discover the “great unread”2 aspects of Tennyson’s work using correspondence analysis, which is an effective statistical and stylometric methodology.
Tabata (2002) employs CA for the distribution of word classes (parts of speech) to investigate the stylistic variation in Charles Dickens’s works. He notes that CA “with word-class variables can be a powerful tool for discerning variations across a large set of texts like Dickens’s oeuvre, which it would be difficult to process by manual handling of the data” (Tabata 2002, 179). Tabata’s study (2002) provides evidence that CA is an effective method in literature research. As Tabata further demonstrates (2004, 112), CA is a useful technique when the target data contain low-frequency content words, namely nouns, verbs, adjectives, and adverbs, that can be used as variables for stylistic research. Tabata (2005, 68) later states that, unlike principal component analysis (PCA) and factor analysis (FA), CA does not require the intermediate processes of computing the correlation and covariance matrices, but analyzes the data directly to generate results. The current study deals with Tennyson and other poets’ low-frequency ModAdvs; thus, CA is utilized as an optimal approach to visualize the hidden links behind the terms and answer the following questions: Do ModAdvs characterize Tennyson (or other poets)? If so, why do these ModAdvs feature in Tennysonian poetry?
The study focuses on Tennyson and eleven other poets, from the Augustan, Romantic, and Victorian eras: Alexander Pope, George Gordon Byron (Lord Byron), John Keats, Percy Bysshe Shelley, Robert Browning, Samuel Taylor Coleridge, William Wordsworth, Algernon Charles Swinburne, Dante Gabriel Rossetti, Matthew Arnold, and Robert Southey. The authors were selected based on Shaw’s (1976) suggestions regarding similarities in style between Tennyson and these authors. In addition, based on Harvey’s (1937) inclusive arrangement of historical English literary authors, this study included four Victorian authors to achieve more comprehensive results. The target text genre of this study is poetry; thus, some writers listed by Shaw and Harvey were excluded because their number of words (i.e., tokens) of poetry comprise fewer than eighty thousand words, for example, Thomas Carlyle and Christina Rossetti. The text data was compiled using only verse texts from the Delphi Poets series3 for the twelve authors. The Delphi Poets Series is based on the original copies of authors’ first published collections. These collections sometimes include both first and revised editions of the same works, for example, Tennyson’s “The Lady of Shallot,” published in 1833 and 1842. Table 1 shows the total tokens for the twelve authors.
The fifteen target ModAdvs are belike, certainly, conceivably, doubtless, doubtlessly, haply, maybe, mayhap, peradventure, perchance, perhaps, possibly, probably, surely and undoubtedly, which primarily express epistemic possibility or uncertainty. They were chosen based on the investigations of Poutsma (1928), Quirk et al. (1985), and Huddleston and Pullum (2002). The raw frequencies and relative frequencies (per million words) of the fifteen ModAdvs are shown in tables 2 and 3, respectively.
This study examines the raw frequency of ModAdvs in poetry, rather than relative frequency, which is more commonly used when total tokens of the target texts differ considerably, as shown in table 1. Several ModAdvs appear infrequently in an author’s entire body of work (table 2). For example, the relative frequencies of doubtless in the works of Byron and Arnold (table 3) appear to have similar values, i.e., 69.57 and 70.46, respectively, but their raw frequencies differ substantially, with values of 26 and 6. Alternatively, the relative frequencies of undoubtedly in the works of Pope, Wordsworth, and Shelley are 2.32, 1.73, and 4.72, respectively, but all have a raw frequency of 1 (table 2). Furthermore, low-frequency words, such as those that appear only once or twice in the entire text, may not increase proportionately with the length of the target text, so normalized values may not accurately reflect their usages. For words that only occur once in an author’s entire oeuvre the vital question is determining where and how the word appears and what words it collocates with; therefore, although the frequency indicates the same value 1, such words will be weighted differently depending on the reasons why they appear scarcely and inconsistently. It is important to mention that CA handles row and column items symmetrically, reducing the dimensionality of the matrix being calculated. This results in outputting both row and column coordinates in the same manner, even when the matrix is transposed. This sets CA apart from other multivariate techniques such as FA, PCA, and hierarchical cluster analysis (HCA). With this feature, CA allows for processing of raw frequency variables without the need for normalizing frequencies based on text size.
Table 4 lists the fifty most frequently appearing words for all twelve authors. Section 4.1 analyzes the fifty most frequent words using CA as a comparison to the CA for ModAdvs. The outcome of the fifty most frequent words can serve as a criterion for evaluating the results of the ModAdvs. This implies that if certain authors’ variables are located in a similar manner in both the results of the fifty most frequent words and the ModAdvs, the results of the CA on ModAdvs may not be solely envisaged as the characteristics of the ModAdvs, but may also be observable in other factors or generally in the corpora.
For the CA analyses and drawing plots of the CA results, this study employed CasualConc 3.0.4 (Imao 2022a). CasualConc is a native concordancer for macOS. The target text files were loaded into CasualConc as each poet’s data (i.e., corpus), and then it counted up word frequencies in each corpus. The word frequencies were then imported to run the “Correspondence” tool, which analyzes the data using CA. Before processing the CA function, the words to be analyzed were filtered so that the data was able to be limited to the optional number of words in the CA pane of CasualConc. Further, in CasualConc’s “Options” window, the analyst can set detailed conditions for its calculation and plotting. This study set the conditions as the results should have been “Scaled: to match the scales of the row item plot and the column item plot,” “Ratio: to match the plot area or the range of X and Y axes,” and “Standardized” to standardize values (Imao 2022b, 252). For its calculations as well as for depicting plots, CasualConc employs corresp in the MASS package of R.
This section presents the results of CA for the top 50 words (section 4.1) and ModAdvs (section 4.2). In the figures shown in this section, the authors’ names were abbreviated as follows: Arnold (ARNO), Browning (BROW), Byron (BYRO), Coleridge (COLE), Rossetti (DROS), Keats (KEAT), Shelley (PSHE), Pope (POPE), Southey (SOUY), Swinburne (SWIN), Tennyson (TENN), and Wordsworth (WORD).
Figures 1 and 2 show the CA results of the 50 most frequent words for all twelve authors, which are listed in table 4. Figure 1 shows the triple-dimension plot (dimensions 1–3) of column coordinates, i.e., the 50 most frequent words, and figure 2 shows the triple-dimension plot (dimensions 1–3) of row coordinates, i.e., the 12 authors.
In figures 1 and 2, the contribution rates for each dimension are indicated in the parentheses following the dimension (D) numbers (1–3). The contribution rate for D1 is 33.69%, for D2 is 23.07%, and for D3 it is 13.43%. The cumulative contribution rate of D1–D3 is 70.19% in both figures 1 and 2.
The column coordinate plot (figure 1) indicate that you (left side of the plot), which,5 and had (center bottom), and thy, then, his, and their (rear right) are furthest from the origin [0.0, 0.0, 0.0] of the box. The further they are located from the center, the more they deviate from the standard in the analyzed data. The proximity of variables explains the similarity of their frequency patterns between variables, i.e., the closer the data points are located to each other, the greater similarity they have in the frequency patterns of the 50 most frequent words. The row coordinates (figure 2) indicate that Browning and Pope, located on the left side and the rear top right of the plot, respectively, are the most distant from the center.
By considering the results shown in figures 1 and 2 collectively, it is evident that the use of you is a characteristic of Browning’s work since the location of the column variable you’s coordinate (figure 1) corresponds to the location of the row variable of Browning’s coordinate (figure 2). The words which and had can be considered features of Wordsworth, Byron, and Shelley. The results shown in figure 2 suggests that the eighteenth-century authors, as well as some Victorians, Swinburne and Arnold, tend to be located 0.0 to 1.0 of D1, and Browning, Tennyson, and Rossetti are placed in the range of –2.5 to 0.0.
Table 5 shows Euclidean distances from TENN to other variables in figure 2 in ascending order from left to right. The distances indicate that Tennyson’s usage of the top 50 words does not diverge significantly from that of the other authors; however, Tennyson’s usage is most comparable to that of Rossetti, followed by Keats and Coleridge compared to the other Victorian poets, i.e., Browning, Arnold and Swinburne. The furthest from Tennyson is Pope (0.303).
Figures 3 and 4 show the row and column plots with the tripartite dimensions of the CA of the fifteen target ModAdvs. The contribution rates of the three dimensions are 42.8% (D1), 25.28% (D2), and 15.11% (D3)6; therefore, 83.19% of the data is drawn in each plot.
In figure 3, the authors are plotted horizontally from the front to the back of the box around 0.0 of D3. As can be seen, Tennyson is obviously an outlier among the twelve poets. The column plot shown in figure 4 explains that 14 of the ModAdvs (excluding maybe) are positioned around 0.0 of D3 while these 14 ModAdvs scatter in the range of –6 to 1 in D2. The elements located on the negative side of D2 in figure 3 represent three Victorian poets—Tennyson, Rossetti, and Browning—and one Romantic poet, Shelley, who was born only seventeen years before Tennyson. The eighteenth-century authors and two Victorian authors, Arnold and Swinburne, are positioned on the positive side of the plot. According to the results shown in figures 3 and 4, doubtlessly, conceivably, undoubtedly, and possibly are characteristics of Browning (from D1), and Tennyson’s use of maybe (from D3) can be considered a feature of his writing. Furthermore, focusing on D3, Tennyson and maybe and mayhap are outliers, and the remaining variables are clustered near the origins of the coordinates.
Although Tennyson is not an outlier in the CA results of the 50 most frequent words (figures 1 and 2), the CA results for the ModAdvs highlight the characteristic use of the ModAdvs maybe and mayhap in Tennyson’s work. The usage of these two ModAdvs is discussed in section 5. coordinate plot of CA results (15 ModAdvs); dimensions 1, 2, and 3
As demonstrated by the CA results, Tennyson’s most characteristic ModAdvs are maybe and mayhap. This section investigates how both maybe and mayhap are used by Tennyson and why they can be considered unique features of Tennyson’s writing.
According to the Oxford English Dictionary online, mayhap (first known to have been used in writing in 1533), which expresses modality under the meaning of perhaps or possibly, has decreased in use, and is currently described as “archaic, literary, and British regional.”7 Even in the nineteenth century, mayhap was favored less than it had been previously; however, several of the eighteenth- and nineteenth-century poets considered in this study (Southey, Browning, and Tennyson), employed this term one, seven, and three times, respectively, in their works (table 2). Poutsma (1928, 35), a linguist who worked relatively close to the Victorian era, states that the words haply, mayhap, conceivably, perchance, and peradventure can only be found in poetry and other forms of higher literary expression. The target texts considered in this study are poems; hence, it is not unexpected that mayhap and other ModAdvs in question, e.g., haply and conceivably, appear in the works I considered.
With regard to Tennyson, Poutsma’s observation agrees in two respects. First, Tennyson, Browning, and Southey used mayhap in their works. However, while Browning and Southey used mayhap in nondialectal poetry, Tennyson used mayhap only in his Lincolnshire dialectal poetry. Among five of Tennyson’s dialectal poems in this study’s data set, mayhap appears in three poems, “Northern Farmer,” “Northern Cobbler,” and “The Spinster’s Sweet-Arts,” and these instances are quoted below (emphasis added). Note that these poems, which are written in the Lincolnshire dialect, are difficult for readers who are unfamiliar with the dialect to understand.
Mowt ‘a beän, mayhap, for she wur a bad un, sheä.
(Tennyson, “Northern Farmer,” 1862; l. 22)
‘Weänt tha?’ she says, an’ mysen I thowt i’ mysen ‘mayhap.’
(Tennyson, “Northern Cobbler,” 1880; l. 68)
Alt’ a-callin’ ma ‘hugly’ mayhap to my faäce,8
(Tennyson, “The Spinster’s Sweet-Arts,” 1885; l. 91)
On the contrary, the language employed by Browning and Southey lacks dialectal elements, as evidenced by the following excerpts:
Save now and then mayhap at Christmas time,
(Southey, “The Wedding,” 1800; l. 19)
By Newman and, mayhap, wise-manned to boot
(Browning, “The Ring and the Book,” 1868; l. 322)
Although these three poets use the word mayhap in their poems, the motivations and purposes of their uses likely differ. For example, Southey’s and Browning’s usage could be a fashion of the time, whereas Tennyson only uses mayhap in his Lincolnshire dialectal poems; therefore, the word contributes to the dialectal character of his poems.
The OED lists the first recorded use of the term maybe as dating from 14009. It adds that although maybe was a common term used by writers in the seventeenth century, it was not widely utilized until the middle of the nineteenth century10. Poutsma (1928, 35) indicates that belike and maybe (as well as may-be or may be) seem to be more or less dialectal or archaic. Tennyson first used maybe in his 1842 poem “St. Simon Stylites.” The period when Tennyson begin to use maybe is in agreement with the description provided by the OED (under Etymology). The aforementioned “literature archaism” (Leech 1969, 13) can be assumed as a factor in the use of maybe in Tennyson’s, Browning’s and Swinburne’s works; but none of the Augustan and Romantic poets use maybe, which is consistent with the OED description.
In this study, only the Victorian poets (Tennyson, Browning, and Swinburne) employ maybe in their poems; the other poets tend to use haply. Browning and Swinburne use maybe six and two times, respectively. In contrast, Tennyson employs maybe relatively frequently (eleven times). The titles of the poems and the frequency of maybe as well as its variation in capitalization are listed in table 6.
Five of seven Tennyson poems are epic and/or narrative poems that provide an account of protagonists’ lives or heroic experiences. These five poems are written in blank verse, which is unrhymed at the end of lines, has a regular meter, and is most often written in iambic pentameter. The adverb maybe scarcely appears at the end of sentences or lines:
Maybe for months, in such blind lethargies,
(Tennyson, “St Simon Stylites,” 1842; l. 101)
He triumphs; maybe11, we shall stand alone:
(Tennyson, “Britons, Guard your Own,” 1852; l. 5)
As can be seen in the quotations from “St Simon Stylites” and “Britons Guard your Own,” although these two poems are not epic/narrative poems, maybe is used to convey the narrators’ epistemic possibility and uncertainty.
In most of the occurrences of maybe, punctuation marks such as dashes (—), commas (,), question marks (?), and semicolons(;), or the conjunctions but, and, and or precede and/or succeed the term, as shown in table 7. This suggests that the instances of maybe are sentence modifiers that mutually interact with the pauses indicated by the punctuation symbols to enhance the speaker’s uncertainty, doubtfulness, or hesitation. In all the examples represented as concordance lines in table 7, the maybe in line 15 by Tennyson is the only adverb that does not modify the sentence it is in. Instead, it modifies the succeeding adjective twain, which means two (here, the noun hours is implied after “twain”).
Its etymological origin allows maybe to differ from other ModAdvs in collocation with modal auxiliaries, e.g., may, might, or can. Maybe and mayhap typically do not collocate with other modal auxiliaries because these words are contractions of the modal auxiliaries “it may be” or “it may hap(pen)” (OED, s.v. maybe and mayhap: Etymology). While maybe collocates only with the auxiliary shall in the texts in question, and mayhap collocates infrequently with auxiliaries, other ModAdvs, for instance, haply, perchance, and perhaps, frequently co-occur with various auxiliaries, for example, can/could, may/mayst/might, must, and shall/shalt/should/shouldst in works of the twelve poets considered in this study. The target ModAdvs primarily contain two syllables: belike, doubtless, haply, maybe, mayhap, perchance, perhaps, and surely. Among them, haply, perchance, perhaps, maybe, and mayhap express a similar level of epistemic possibility or uncertainty; thus, those could be alternatives to use of maybe, even though Tennyson opts to use maybe. In choosing among options for maybe, phonetic and prosodic factors can be a consideration; that is, the collocations with auxiliaries expand or narrow the divergences of the sound/rhythm patterns and the adaptations of alliteration or internal rhyme observed in the following lines:
Or, maybe, I myself, my bride once seen
(Tennyson, The Princess, 1847: l. 72)
The spreading blue bells: it may haply mourn / That such fair clusters should be rudely torn
(Keats, “I Stood Tip-Toe upon a Little Hill,” 1817: ll. 43–44)
In these lines, even though maybe and haply have similar meaning, the alliteration, rhythm, and rhyme, which show poetry’s personality, would be compromised if these terms were replaced. Further, if maybe had replaced haply in Keats, the preceding auxiliary may would not have been used and the meter wouldn’t have worked without further rewriting; therefore, maybe and haply were not necessarily irreplaceable in the context of poetry, albeit they behave syntactically and semantically alike.
This paper has investigated Tennyson’s characteristic use of modal adverbs using the quantitative CA method, and I compared his use of the modal adverbs maybe and mayhap to that of eleven other Augustan, Romantic, and Victorian poets. Moreover, I discussed why Tennyson preferred the use of maybe and mayhap in his poems. Even though this analysis was limited to only fifteen modal adverbs and the work of eleven reference poets, CA revealed some new aspects of Tennyson’s style and facilitated the examination of verse text to also reveal an “unread” aspect of literary texts. In the future, I plan to perform a detailed qualitative analysis of additional target terms and reference poems.
Acknowledgment
This paper is partially based on a presentation by the author at JADH 2020, “A New Decade in Digital Scholarship: Microcosms and Hubs,” November 20–22, online12. I would like to thank the chairs and audience members for the useful discussion at the conference.
This work was supported by the Japan Science and Technology Agency Support for Pioneering Research Initiated by Next Generation [grant number JPMJSP2138].
The data analysis file and all annotator data files are not openly available.
The term “great unread” is defined by Moretti (2013, 45) in reference to a concept originally proposed by Margaret Cohen.
Ditto.
Collections for each author were purchased as eBooks, and then converted with permission to plain-text data for the purpose of the author’s research. URL for each eBook of author is listed below.
Pope: https://www.delphiclassics.com/shop/alexander-pope-2/,
Wordsworth: https://www.delphiclassics.com/shop/william-wordsworth-2/,
Coleridge: https://www.delphiclassics.com/shop/samuel-taylor-coleridge-2/,
Southey: https://www.delphiclassics.com/shop/robert-southey-2/,
Byron: https://www.delphiclassics.com/shop/lord-byron-2/,
Shelley: https://www.delphiclassics.com/shop/percy-bysshe-shelley-2/,
Keats: https://www.delphiclassics.com/shop/john-keats-2/
Tennyson: https://www.delphiclassics.com/shop/alfred-lord-tennyson-2/,
Browning: https://www.delphiclassics.com/shop/robert-browning-2/,
Arnold: https://www.delphiclassics.com/shop/matthew-arnold-2/,
Rossetti: https://www.delphiclassics.com/shop/dante-gabriel-rossetti-2/,
Swinburne: https://www.delphiclassics.com/shop/algernon-charles-swinburne-2/
Frequencies were normalized per one million words.
Words with multiple meanings and syntactic functions, such as which, when, and what, were regarded as one word each in this analysis, regardless of whether they appeared as interrogatives or relative pronouns. For which, 10% of its frequencies were randomly extracted from each author and then tallied to determine whether it was used as an interrogative or a relative pronoun. The proportion of its usage as an interrogative in each author's dataset was smaller in comparison to its use as a relative pronoun across all authors, as demonstrated in Table 8.
The figures 3 and 4 represent the three most significant dimensions based on the contribution rates.
Oxford English Dictionary, “mayhap” (Meaning & use), accessed December 23, 2022, https://www.oed.com/dictionary/mayhap_adv?tab=meaning_and_use
The alteration of spelling, terminology, or other content in lines often occurs in poetry. This 91st line from “The Spinster’s Sweet-Arts” is not an exception. This article employs and follows the wordings in Delphi Poets series.
Oxford English Dictionary, “maybe” (Meaning & use), accessed December 23, 2022, https://www.oed.com/dictionary/maybe_adv?tab=meaning_and_use#37912745
Oxford English Dictionary, “maybe” (Etymology), accessed December 23, 2022, https://www.oed.com/dictionary/maybe_adv?tab=etymology#37913193
There are versions of this poem that spell the word as “may be” (auxiliary may and be verb); however, this paper employs the spelling used in the Delphi“Great Poets” series.
The proceeding of the presentation is available in the following URL: https://jadh2020.lang.osaka-u.ac.jp/programme/jadh2020proceedings.pdf
