Style and Authorship in a Classic of Popular Culture: Henry Livingston and "The Night Before Christmas".
No experienced reader of poetry who had worked their way through the extant verse of both Livingston and Moore could fail to recognize that their characteristics are dissimilar. Livingston's output is warm-hearted, celebratory, and imaginative. It has verve and humor. The shorter pieces have an easy lyric grace. Livingston's poetic personality is playful and whimsical. Moore, in contrast, is a moralist, pedagogue, and satirist. He is inclined to preach. His poems are often clogged with earnest cogitation. Whereas Livingston's lines trip off the tongue, Moore's trip up the tongue. But of course attribution studies must turn such subjective literary-critical impressions into objectively quantifiable data.
"The Night Before Christmas" deserves its fame. The imagined encounter with an elfin St. Nick, his tiny reindeer, and miniature airborne sleigh is told with verve and considerable narrative skill, riveting attention from beginning to end. Energized by a string of active verbs, the poem is crammed with sensory detail. A substantial article might be devoted to analysis of its organization, as the speaking voice modulates--rising and falling, speeding up and slowing down--through a variety of tones and emotions (expectation, surprise, excitement, wonder, amusement, and joy) and of the range of rhetorical figures and poetic devices employed, right through to the chiasmic structure (abba) of the concluding line, "Happy Christmas to all, and to all a good night." But distinguishing between authors requires a more mundane approach.
In a recent book, I attempted a thorough reassessment of the arguments in favor of Moore's or Livingston's authorship and described statistical tests that distinguish between the acknowledged poetry of the two claimants (Jackson, Who Wrote). When these were applied to "The Night Before Christmas," it turned out to be consistently associated with Livingston, not Moore. The evidence as a whole seemed to warrant the reattribution of the famous poem to Livingston.
Both Moore and Livingston had produced a sufficient amount of verse for the purposes of differentiating between their styles. Since no other candidate had ever been, or was ever likely to be, proposed as author of "The Night Before Christmas," it was necessary only to search for features that in combination discriminated between one man's work and the other's. The results were unequivocal. It might, however, be claimed--though not plausibly, as I tried to show--that they were a consequence of genre, rather than of authorship: Moore aspired to join a poetic mainstream, whereas Livingston belonged to a more popular rhyming tradition, associated at the time with fugitive publication in newspapers and magazines. It has therefore seemed worth conducting a follow-up investigation comparing Livingston's poems with those of a range of his peers and seeing whether "lhe Night Before Christmas" more comfortably fits within the Livingston or Non-Livingston body of verse.
The core corpus of poems incontestably Livingston's is listed by Jackson (Who Wrote 164-68). It consists of manuscript items in the poet's own handwriting, either in a bound collection of his own or inscribed in his daughter Jane's "poetry book"; items attributed and owned by other descendants; and items published under Livingston's pseudonym "R" in newspapers and journals to which he is known to have contributed.
A comparative Non-Livingston corpus--matched as closely as possible with Livingston's for overall size, date, and kinds of subject matter--was compiled by Mary Van Deusen from newspapers and journals of the appropriate period and in most of which Livingston's verse had also been published: the American Magazine, New York Magazine, Northern Whig Poughkeepsie Journal, and Weekly Museum. The verses are mainly fresh compositions, but a small minority, by earlier poets, are reprinted in such places as the Poughkeepsie Journal's "Poets' Corner" as favorites of contemporary readers; these latter have been retained as indicative of some of the models in vogue. A complete list of the titles, the given authors or pen names, and the sources (with journals and dates) is provided in Appendix 1.
Especially effective discrimination between Livingston and Moore had been achieved through counts of the rates of use of the most commonly occurring words. High-frequency words or "function words"--"the," "to," "with," "for," "of," and the like--belong in almost every modern attribution scholar's armory (Burrows; Juola). They form the framework on which sentences, with their "content" words (nouns, verbs, adjectives, adverbs) are built, and so their rates of use register preferences in the deployment of grammar and syntax. A graph showed "The Night Before Christmas" falling outside the range for Moore poems and close to the middle of Livingston's range (Jackson, Who Wrote 133). A similar methodology is here employed in an attempt to distinguish between the Livingston and Non-Livingston corpora.
Paul Kosinski's computer program created inventories of the frequencies of use (number of tokens) of every word (type) in each corpus--Livingston's of 1,893 lines and 13,074 words as tokens and Non-Livingston of 1,872 lines and 13,168 words as tokens--and calculated the rates of use for each type as a percentage of the total number of tokens. In each inventory, word-types were ordered according to frequency and, when of equal frequency, alphabetically. Rates of the one hundred and ten most frequentiy occurring words in either corpus were then compared with the rates of occurrence in the other corpus, and those that occurred at rates at least 1.2 times greater in one or other corpus were selected as "Livingston-favored" or "Non-Livingston favored" words. If the one hundred and tenth greatest frequency was shared by several words, the alphabetical order was used to determine the cut-off point. Since many of Livingston's pieces are restricted to manuscript, such variant spellings as "'tis" or "tis," "oh" or "o," "and" or "&," "though" or "tho"' or "tho," and "every" or "ev'ry" were combined as a single word. Counts amalgamated those for capitalized and non-capitalized words. There were forty-six Livingston-favored words, forty five Non-Livingston favored words. They are listed in Appendix 2. Although function words dominate the lists, a few content words also qualify for inclusion.
In many authorship studies, rates of use in works by Author A and Author B, or by Author A and a comparative group of Authors B-Z, are plotted on two separate axes, vertical and horizontal, but for this study, as for the earlier investigation of Livingston versus Moore, one simple figure was calculated--the number of Livingston-favored words as a percentage of the total of Livingston-favored plus Non-Livingston favored words. Four Livingston poems and four Non-Livingston poems contained fewer than fifteen of the test words and have been discarded from Table 1 (in Appendix 3) and Figure 1, which show the results.
The two corpora clearly form two different populations. Livingston's has a mean of 71.673, a standard deviation of 11.571, and a range of 40.000-95.122, with "Anne" at the lower end an obvious outlier, the next lowest score being 52.632. The Non-Livingston comparative group has a mean of 48.225, a standard deviation of 11.137, and a range of 22.222-78.261, with "Epitaph on a Sailor" at the upper limit an obvious outlier, the next highest score being 71.429. Livingston's "Anne" ("To My Little Niece Anne Duyckinck") is very short, consisting of only twelve lines and sixty-two words altogether. "Epitaph on a Sailor" sustains a series of nautical metaphors for sixteen lines: the dogged ingenuity with which it does so verges on the comic. Livingston used nautical imagery in his epitaphs "On the Late Mr. Gilbert Cortland," "To the Memory of Sarah Livingston," and "Catherine Livingston Breese," but he handles the limited number of such metaphors with tact and without strain. Thirty-four years after the anonymous publication of "Epitaph on a Sailor" in the Poughkeepsie Journal, a garbled version was used on the tombstone of Reuben Chase, a Nantucket naval hero of the Revolutionary War, who served as a midshipman under John Paul Jones on the Ranger and died on July 23, 1824 (Anon, "Reuben Chase," "Epitaph").
This means that only one anomalous Non-Livingston poem, out of fifty eight, scores more highly on its percentage of Livingston-favored words than "The Night Before Christmas," with 75.676. On this test, "The Night Before Christmas" belongs within the Livingston corpus.
In Figure 1 percentages of Livingston-favored words are rounded off to the nearest whole number and the numbers of Livingston and Non-Livingston poems that fall into the various ranges of percentages (20-24, 25-29, 30-34, etc.) are displayed in bar-graph form.
An innovative test of "phoneme pairs" proved efficacious in discriminating between the poetic corpora of Livingston and Moore (Jackson, Who Wrote 49-63). Texts were transcribed into the phonetic code known as Arpabet, which was developed by the Advanced Research Projects Agency and designed for General American English ('Arpabet"). It is used in Carnegie Mellon University's CMU Pronouncing Dictionary, available online. A phoneme is a unit of significant sound in a specific language. In Arpabet the word "stockings" consists of the seven phonemes S-T-AA-K-IH-NG-Z. Arpabet thus has the advantage over the International Phonetic Alphabet of employing only familiar letters of the English alphabet, though in its own manner. Phoneme pairs were defined as consisting of the last phoneme of one word and the first phoneme of the next word within a single verse line: a line's final phoneme could not create a phoneme pair with the initial phoneme of the following line. The phrase "not even a mouse," transcribed as N-AA-TIY-V-IH-N AH M-AW-S, yielded the phoneme pairs T/IY, N/AH, and AH/M. Kosinski's computer program counted the frequencies--within individual poems by Moore and Livingston and within their corpora as wholes--of all the different phoneme pairs.
The rates of occurrence of different phoneme pairs are good indicators of authorship, particularly of poetry, because they register a combination of (a) choice of words and (b) ways of putting them together. A poet's preferences among them are governed by both sound and sense. Livingston's verse is more euphonious and flows more freely than Moore's, so that he uses more phonetic pairs of which one element is a vowel sound than does Moore, whose most favored phoneme pairs tend to have consonants as each element. Again, a graph of results for totals of Livingston-favored phoneme pairs, expressed as a percentage of Livingston-favored plus Moore-favored pairs showed "The Night Before Christmas" falling outside the range for Moore's poems and close to the mean score for Livingston's (Jackson, Who Wrote 132).
For the present study, the incidence of different phoneme pairs is compared in the Livingston and Non-Livingston corpora. The question is again whether the two populations are statistically distinguishable and, if so, to which does "The Night Before Christmas" most naturally belong. Each of the hundred most frequently occurring phoneme pairs in either corpus was checked by chi-square to determine whether it was used at a higher rate within one or other overall corpus. A 5 percent level of significance was adopted (p < 0.05). The following sets of phoneme pairs qualified:
Livingston-favored: AH/N AY/HH AY/L D/AH DH/DH D/N IY/AH IY/K N/ AH T/AH Z/IH
Non-Livingston-favored: AY/DH L/B OW/M S/DH T/DH Z/DH Z/S
Remarkably, Livingston's preference for pairs that include at least one vowel sound distinguishes his verse not only from Moore's, but from that of his fellow contributors to the newspapers and popular journals of the time. Nine of Livingston's eleven pairs in the above list contain a vowel, as do only two of the seven Non-Livingston pairs. Italian is considered a "musical" language largely because words end in vowels, and the statistics support our sense that Livingston's typical poetic style is unusually melodious and nimble.
For every poem the number of instances of Livingston-favored phoneme pairs was computed as a percentage of the total number of instances of all eighteen pairs. In each set, twenty poems contained fewer than ten of the test pairs, and so are eliminated from the results that are given in Table 2 (in Appendix 3) and graphed in Figure 2, which is constructed in the same way as Figure 1. Results for the forty poems that afford so few relevant data vary wildly. Obviously a paucity of data increases the degree of purely random variation. Discarded Livingston poems range from 25% Livingston-favored words for "Catherine Livingston Breese," which has only four test pairs, to 100% for "Without Distinction," which has only one test pair; the four-line "Catherine Breese Livingston" has none. The overall percentage for the twenty discarded Livingston poems combined as a single unit is 63.964. Discarded Non-Livingston poems range from 0% for "Old Man's Advice" (with seven test pairs), "Eulogy on Greene" (with nine), "Fatal Beverage" (with one), and "Sonnet to Hope" (with two) to 83.333 for "Retirement" (with six). The overall percentage for the twenty discarded Non-Livingston poems is 40.152.
Of prime interest, however, are the results for poems with at least ten test pairs. The forty-four Livingston poems that qualify have percentages of Livingston-favored pairs ranging from 50.000 to 92.857, a mean of 69.356, and a standard deviation of 11.602. The forty-two Non-Livingston poems that qualify have percentages ranging from 13.636 to 78.582, a mean of 43.719, and a standard deviation of 11.875. The poem at the top end of the Non-Livingston range, "Wedding Ring" at 78.582, is the only one with a higher score than "The Night Before Christmas" with 77.551. "Wedding Ring" is an outlier among the Non-Livingston poems that qualify, the next highest score being 66.667 for "Philadelphia." Moreover, "Wedding Ring" has only fourteen test pairs, so that if we were to set our cut-off point at fifteen (as for test words), rather than ten, "The Night Before Christmas" (which has forty-nine test pairs) would be well clear of any Non-Livingston poem. The lower cut-off point was adopted because there are far fewer test pairs than test words. But again it is clear that "The Night Before Christmas" belongs with the Livingston poems.
"Wedding Ring" scored below the Non-Livingston mean for its percentage of Livingston-favored words. "Epitaph on a Sailor," which topped the Non-Livingston percentages for Livingston-favored words, had only six test pairs, so was unsuited to the analysis of Phoneme Pairs. In combination the Common Words and the Phoneme Pairs establish "The Night Before Christmas" as more Livingston-like than any of the poems in the Non-Livingston group.
We may conclude with a brief summary and discussion of our findings. The two tests that had, in a previous study, been found to discriminate most effectively between Livingston's poems and Moore's poems were in the present study found to discriminate also between Livingston's poems and a body of Non-Livingston poems by a variety of poets who published in the same or similar journals as did Livingston during the same period. The two tests--of Common Words and Phoneme Pairs--are almost entirely independent of each other. Forty-two Non-Livingston poems qualify for inclusion in both Table 1 (showing each poem's percentage of Livingston-favored common words) and Table 2 (showing each poem's percentage of Livingston-favored phoneme pairs). The rank-order correlation between the two sets of percentages is very low: r = 0.1577,42 d.f., p = 0.3186 (two-tailed). Statistically, this is completely insignificant.
It might be suggested that the performance of "The Night Before Christmas" on this test is, to a significant degree, the result of its having been composed in anapests (ti-ti-turn), as were about one-third of Livingston's poems (those listed by Jackson, Who Wrote 33)--that meter significantly determined the selection of high-frequency words and phoneme pairs. But we can reject this suggestion for the following reasons. Ten Non-Livingston poems in our Tables 1 and 2 (Appendix 3) are anapestic or mainly so. They are identified by asterisks in Appendix 1. Their total percentage of Livingston-favored words is 51.938, slightly above the total for all Non-Livingston poems (46.738) but far short of the total for Livingston poems (73.158) and for "The Night Before Christmas" (76.676). Similarly the total percentage of Livingston-favored phoneme pairs in anapestic Non-Livingston poems (including "Lydia" and "To Miss E.W."--with too few data to qualify for Table 2) is 44.737, slightly above the total for all Non-Livingston poems (43.387), but far short of the total for Livingston poems (69.524) and for "The Night Before Christmas" (77.551). Among the Livingston poems also, we find little difference between the anapestic pieces and the rest. The percentage of Livingston-favored words in his anapestic poems is 75.454, only slightly higher than for all Livingston's poems (73.158), while the percentage of Livingston-favored phoneme pairs in his anapestic poems is 68.506, slightly lower than for all Livingston's poems (69.524).
The evidence that consistently placed "The Night Before Christmas" among Livingston's, rather than Moore's, poems on tests designed to differentiate the works of the two authors seemed impossible to reconcile with Moore's authorship of the famous Christmas poem:
No doubt a writer may have a flash of inspiration that enables him or her to achieve a unique literary success that seems in a sense "out of character." But to postulate that Moore enjoyed such a serendipitous episode will hardly account for the nature of "The Night Before Christmas." We would still have to explain why Moore, on the solitary occasion in which he created something that was to catch the world's imagination, slipped into a style that was not only utterly atypical of his own verse but utterly typical of the verse of the very man who, according to his descendants, was the true author. (Jackson, Who Wrote 134)
The results reported in the present article further strengthen this argument. "The Night Before Christmas," which had proved in an earlier study to be more Livingston-like than any poem by Moore that survives in print or manuscript, also turns out to be more Livingston-like than any poem in a large sample by a wide range of Livingston's contemporaries. The tests producing such an outcome are of elements of composition not readily subject to imitation. The rates of use of high-frequency words and of phoneme pairs are largely outside a writer's conscious control.
A previously overlooked piece of "external evidence" for Moore's claim to "The Night Before Christmas" has, however, recently been discovered by Scott Norsworthy: in correspondence to the New York American of 1 March 1844 Moore, correcting a misattribution of the poem to the artist Joseph Wood, unequivocally asserted his authorship (Norsworthy). This was just before he included the Christmas piece in his Poems of that year. But there are oddities about the timing of Moore's assertion and inaccuracies in his letter, which, taken together, arouse suspicion (Jackson Response), while the documentary evidence crediting Livingston with the poem cannot lightly be dismissed. The internal evidence associating "The Night Before Christmas" with Livingston may not be sufficient to carry the day, but it presents a daunting challenge to Moore's supporters: it cannot easily be explained away.
APPENDIX I: NON-LIVINGSTON CORPUS
In the list below, between semi-colons are given the title of the poem (sometimes slightly shortened), the attribution provided in the journal from which the text was taken, an acronym of the journal tide, the date of the relevant issue. The authors' names are mostly pen-names. In some cases the attribution is merely to another journal: poems were often recycled. Asterisks indicate pieces wholly or mainly anapestic. The acronyms identify the following journals: AM American Magazine, NYM New York Magazine, NW Northern Whig PJ Poughkeepsie Journal, and WkM Weekly Museum.
"Aleon," Ella, NYM, March 1791; "Almeria," Imona, NYM, January 1791; "Ancient Poetry," T. Carew, PJ, 9 April 1794; "Autumn Elegy," Emma, PJ, 11 December 1804; "A Widow Lady," A Customer, PJ, 10 April, 1793; *"Banning Ramsay," Philip Freneau, 6 December 1786; "Belle's Progress," From Norfolk Beacon, PJ, 15 May 1822; "Broken Flute," D, NYM, March 1791; *"Brown Cow," A New-York Farmer, NYM, January 1791; "Cleora," Ann Eliza Bleecker, NYM, May 1791; * "Columbia," S of NJ, PJ, 24 April 1822; "Cornelia Remsen," Author of Juvenis, NYM, February 1791; "Death of Hillard," Ray, PJ, 4 December 1804; "Delia Crusca," Mrs. Morton, NYM, March 1791; "Dutch People," Anon., PJ, 18 December 1804; "Epitaph on a Sailor," Anon., PJ, 24 January 1787; "Epithalamium," Philo, PJ, 9 May 1787; "Eulogy on Greene," William Hillhouse, PJ, 6 December 1786; "Fatal Beverage," Julian Prefect of Egypt, PJ, 19 March 1794; *"Fox and Cat," John Cunningham, WkM, 12 July 1794; "Friendship," From Repertory, PJ, 2 December 1806; "Full Blown Rose," P, PJ, 3 April 1822; "Galley Slave," Author of Johnson, PJ, 22 December 1785; "General Knox," From Boston Gazette, PJ, 16 December 1806; "Hail Boreas," A Refugee, PJ, 20 December 1803; "Happiness," Anne Steele, PJ, 27 June 1787; "Indian Eclogue," From Chronicle of Freedom, PJ, 16 May 1787; "Juliet," Strephon, NYM, May 1791; "Laura," Peter Pindar, NYM, June 1791; "Logan's Triumphs," W.D., NYM, January 1791; *"Lover's Vows," From Literary Gazette, PJ, 6 February 1822; *"Lydia," Leander, AM, December 1787; "Major Wyllys," A Friend, NYM, February 1791; "Mercy," Selleck Osborn, PJ, 30 December 1807; "Miniature Profile," Julia, NYM, February 1791; "Miranda's Birthday," Ferdinand, NYM, March 1791; "Miss Anna D-nd-s," Ella, NYM, April 1791; "Mr. Brakenridge," From Chronicle of Freedom, PJ, 18 April 1787; "Ode to Learning," M.D., PJ, 17 April 1793; "Old Man's Advice," Nestor, PJ, 18 July 1787; "On Dr. Trapp," Dr. Trapp, PJ, 17 October 1785; "On the President," S.L., NYM, January 1791; "Philadelphia," From Chronicle of Freedom, PJ, 6 June 1787; "Present Hour," John Trumbull, PJ, 20 September 1786; "Rebus," Corydon, NYM, March 1791; "Retirement," Dr. Brooks, PJ, 20 June 1787; *"Rich and Ambitious," L, PJ, 9 February 1786; "Sailor Boy," From New York Daily Advertiser, PJ, 10 December 1805; "Serenade," Professor Frisbie, PJ, 29 May 1822; "Setting Sun," Mrs. Pearson, NYM, April 1791; "Situation in France," Mareia, NYM, January 1791; "Sonnet to Hope," Helen Maria Williams, NYM, January 1791; "Striking a Fly," Damon, AM, December 1787; *"Sylph," Florio, NW, 27 April 1819; *"To Cymon," Ella, NYM, January 1791; "To Florio," Hortentius, NW, 7 December 1819; *"To Miss E.W.," J.B.C., NYM, January 1791; "Troubled Ocean's Face," Eliza, PJ, 13 December 1786; "Vision of Franklin," From Vision of Columbus, PJ, 2 May 1787; "Wedding Ring," George Doane, PJ, 4 July 1827; "Winter's Walk," Dr. S. Johnson, NYM, January 1791; "Year Grows Old," From Middlesex Gazette, PJ, 13 December 1786
APPENDIX 2: TEST WORDS
Livingston-favored words: a, all, and/&, around, at, could, day, dear, do, down, each, ever, found, good, had, happy, have, high, I, if, its, lay, little, lovely, man, meet, name, new, on, or, our, peace, plain, rose, than, till, upon, was, we, what, where, who, whose, will, would, you. (46 words)
Non-Livingston-favored words: are, away, be, before, bosom, breast, bright, care, come, death, fair, fame, fly, for, form, heart, her, here, how, let, like, men, mind, must, no, o/oh, round, see, shall, she, sing, some, soul, still, sweet, thee, they, this, those, thou, thy, tis/'tis, were, yet, your. (45 words)
APPENDIX 3: TABLES
Table 1: Common words more favored in Livingston or Non-Livingston corpus (a) Number of (b) Number of Livingston- Non-Livingston- Favored Words Favored Words (as tokens) (as tokens) Livingston poems Acknowledgement 47 9 Acrostic 21 18 Alcmena 40 14 American Eagle 115 13 Anne 8 12 Apollo 42 17 Arabella 10 9 Bats (Fable) 32 16 Beekman 77 41 Belle 31 10 Carrier 1787 75 20 Carrier 1803 124 58 Carrier 1819 120 26 Careless 30 7 Catherine L Breese 13 6 Catherine Sleeping 22 24 Country 54 10 Crane and Fox 88 36 Dance 57 23 Death of Sarah 17 12 Deity 30 15 Dialogue 21 11 Easter 19 16 Eliza Hughes 9 6 Fly 19 9 Frog King 42 11 Frontier Song 13 4 Frosts (Habakkuk) 14 5 Gentleman 53 18 German Spa 13 10 Gilbert Courtland 19 4 God is Love 39 2 Henry Welles L 13 5 Hero 64 15 Hezekiah 22 9 Hogs 45 13 Isaiah 24 13 Joanna 17 15 Job 19 3 Lo from the East 21 5 Marriage 19 12 Marriage Tax 49 11 Midas 28 3 Monarchs 22 4 Montgomery 20 10 Past is the Hours 21 17 Procession 23 12 Queen 25 2 Rispah 25 8 Sages 46 11 Scots Wha Hae 17 5 Settlement 24 16 Sisters 31 5 Spadille 37 20 Tappen 27 9 Timmy 28 17 To Miss 24 10 Valentine 29 8 Vine 66 28 War 58 10 Wren 17 10 Totals: L = 2171 Non-L = 798 Mean of percentages for individual poems: Non Livingston poems A Widow Lady 20 23 Aleon 20 20 Almeria 16 29 Ancient Poetry 24 20 Autumn Elegy 33 57 Banning Ramsay 30 16 Belles Progress 34 24 Broken Flute 47 73 Brown Cow 40 37 Cleora 35 45 Columbia 8 28 Cornelia Remsen 29 44 Death of Hillard 17 13 Delia Crusca 15 16 Dutch People 21 31 Epitaph on a Sailor 18 5 Epithalamium 11 14 Eulogy on Greene 13 11 Fox and Cat 34 22 Friendship 42 49 Full Blown Rose 32 26 General Knox 7 10 Hail Boreas 8 12 Happiness 14 8 Indian Eclogue 29 21 Juliet 22 41 Laura 11 19 Logan's Triumphs 30 61 Lover's Vows 22 11 Lydia 26 25 Major Wyllys 47 29 Mercy 25 10 Miniature Profiles 14 16 Miranda's Brithday 54 76 Miss Anna D-nd-s 43 51 Mr. Brakenridge 19 23 Ode to Learning 17 26 Old Man's Advice 15 15 On Dr. Trapp 11 9 On the President 6 11 Philadelphia 64 85 Present Hour 25 24 Retirement 11 13 Rich and Ambitious 24 21 Sailor Boy 28 14 Serenade 34 65 Setting Sun 13 9 Situation in France 29 44 Sonnet to Hope 8 20 Striking a Fly 10 15 Sylph 26 35 To Cymon 43 23 To Florio 15 16 To Miss E.W. 15 30 Vision of Franklin 10 8 Wedding Ring 20 26 Winter's Walk 13 13 Year Grows Old H 13 Totals: L = 1361 Non-L = 1551 Mean of percentages for individual poems: Night Before Christmas L = 112 Non-L = 36 Column "a" Figures as Percentage of Columns "a" + "b" Livingston poems Acknowledgement 83.929 Acrostic 53.846 Alcmena 74.074 American Eagle 89.844 Anne 40.000 Apollo 71.186 Arabella 52.632 Bats (Fable) 66.667 Beekman 65.254 Belle 75.610 Carrier 1787 78.947 Carrier 1803 68.132 Carrier 1819 82.192 Careless 81.081 Catherine L Breese 68.421 Catherine Sleeping 47.826 Country 84.375 Crane and Fox 70.968 Dance 71.250 Death of Sarah 58.621 Deity 66.667 Dialogue 65.625 Easter 54.286 Eliza Hughes 60.000 Fly 67.857 Frog King 79.245 Frontier Song 76.471 Frosts (Habakkuk) 73.684 Gentleman 74.648 German Spa 56.522 Gilbert Courtland 82.609 God is Love 95.122 Henry Welles L 72.222 Hero 81.013 Hezekiah 70.968 Hogs 77.586 Isaiah 64.865 Joanna 53.125 Job 86.364 Lo from the East 80.769 Marriage 61.290 Marriage Tax 81.667 Midas 90.323 Monarchs 84.615 Montgomery 66.667 Past is the Hours 55.263 Procession 65.714 Queen 92.593 Rispah 75.758 Sages 80.702 Scots Wha Hae 77.273 Settlement 60.000 Sisters 86.111 Spadille 64.912 Tappen 75.000 Timmy 62.222 To Miss 70.588 Valentine 78.378 Vine 70.213 War 85.294 Wren 62.963 Totals: Percent L-words 73.158 Mean of percentages 71.673 for individual poems: Non Livingston poems A Widow Lady 46.512 Aleon 50.000 Almeria 35.556 Ancient Poetry 54.545 Autumn Elegy 36.667 Banning Ramsay 65.217 Belles Progress 58.621 Broken Flute 39.167 Brown Cow 51.948 Cleora 43.750 Columbia 22.222 Cornelia Remsen 39.726 Death of Hillard 56.667 Delia Crusca 48.387 Dutch People 40.385 Epitaph on a Sailor 78.261 Epithalamium 44.000 Eulogy on Greene 54.167 Fox and Cat 60.714 Friendship 46.154 Full Blown Rose 55.172 General Knox 41.176 Hail Boreas 40.000 Happiness 63.636 Indian Eclogue 58.000 Juliet 34.921 Laura 36.667 Logan's Triumphs 32.967 Lover's Vows 66.667 Lydia 50.980 Major Wyllys 61.842 Mercy 71.429 Miniature Profiles 46.667 Miranda's Brithday 41.538 Miss Anna D-nd-s 45.745 Mr. Brakenridge 45.238 Ode to Learning 39.535 Old Man's Advice 50.000 On Dr. Trapp 55.000 On the President 35.294 Philadelphia 42.953 Present Hour 51.020 Retirement 45.833 Rich and Ambitious 53.333 Sailor Boy 66.667 Serenade 34.343 Setting Sun 59.091 Situation in France 39.726 Sonnet to Hope 28.571 Striking a Fly 40.000 Sylph 42.623 To Cymon 65.152 To Florio 48.387 To Miss E.W. 33.333 Vision of Franklin 55.556 Wedding Ring 43.478 Winter's Walk 50.000 Year Grows Old 51.852 Totals: Percent L-words 46.738 Mean of percentages 48.225 for individual poems: Night Before Christmas Percent L-words 76.676 Table 2: Phoneme pairs more favored in Livingston or Non-Livingston corpus (a) Number of (b) Number of Occurrences of Occurrences of Livingston- Non-Livingston- Favored Phoneme Favored Phoneme Pairs Pairs Livingston poems Acknowledgement 15 3 Alcmena 11 10 American Eagle 27 9 Apollo 17 10 Bats (Fable) 14 8 Beekman 25 4 Carrier 1787 23 6 Carrier 1803 35 19 Carrier 1819 33 21 Careless 9 1 Catherine Sleeping 7 4 Country 25 7 Crane and Fox 17 10 Dance 24 8 Deity 9 3 Dialogue 10 2 Easter 14 3 Frog King 12 3 Frontier Song 7 3 Gentleman 14 5 God is Love 19 5 Henry Welles L 13 5 Hero 22 4 Hogs 15 5 Isaiah 7 6 Joanna 9 3 Job 6 5 Lo from the East 16 7 Marriage 9 9 Marriage Tax 9 6 Midas 13 4 Monarchs 17 6 Montgomery 8 2 Past is the Hour 7 7 Processions 6 6 Queen 6 6 Sages 19 8 Scots Wha Hae 13 1 Settlement 17 10 Sisters 8 8 Spadille 10 4 Timmy 9 6 Vine 22 10 War 21 12 Wren 8 4 Totals: L = 657 Non-L = 288 Mean of percentages for individual poems: Non-Livingston poems A Widow Lady 6 7 Aleon 6 6 Almeria 5 10 Ancient Poetry 3 7 Autumn Elegy 3 19 Banning Ramsay 10 12 Belle's Progress 5 8 Broken Flute 16 32 Brown Cow 11 12 Cleora 5 16 Columbia 7 9 Cornelia Remsen 16 11 Death of Hillard 3 7 Delia Crusca 7 8 Dutch People 12 10 Epithalamium 6 8 Fox and Cat 14 14 Friendship 9 14 Full Blown Rose 16 23 Indian Eclogue 6 7 Juliet 5 11 Laura 5 6 Logan's Triumphs 15 15 Lover's Vows 3 17 Major Wyllys 16 17 Mercy 8 5 Miniature Profile 4 7 Miranda's Birthday 22 38 Miss Anna D-nd-s 17 20 Mr. Brakenridge 6 6 Ode to Learning 7 10 Philadelphia 12 6 Present Hour 8 7 Rich and Ambitious 7 8 Sailor Boy 11 6 Serenade 6 12 Situation in France 8 10 Sylph 20 17 To Cymon 8 6 To Florio 6 9 Vision of Franklin 3 12 Wedding Ring 11 3 Totals: L = 374 Non-L = 488 Mean percentage for individual poems: Night Before Christmas L = 38 Non-L = 11 Column "a" Figures as Percentage of Columns "a" + "b" Livingston poems Acknowledgement 83.333 Alcmena 52.381 American Eagle 75.000 Apollo 62.963 Bats (Fable) 63.636 Beekman 86.207 Carrier 1787 79.310 Carrier 1803 64.815 Carrier 1819 61.111 Careless 90.000 Catherine Sleeping 63.636 Country 78.125 Crane and Fox 62.963 Dance 75.000 Deity 75.000 Dialogue 83.333 Easter 82.353 Frog King 80.000 Frontier Song 70.000 Gentleman 73.684 God is Love 79.167 Henry Welles L 72.222 Hero 84.615 Hogs 75.000 Isaiah 53.846 Joanna 75.000 Job 54.545 Lo from the East 69.565 Marriage 50.000 Marriage Tax 60.000 Midas 76.471 Monarchs 73.913 Montgomery 80.000 Past is the Hour 50.000 Processions 50.000 Queen 50.000 Sages 70.370 Scots Wha Hae 92.857 Settlement 62.963 Sisters 50.000 Spadille 71.429 Timmy 60.000 Vine 68.750 War 63.636 Wren 66.667 Totals: Percent L-pairs 69.524 Mean of percentages 69.356 for individual poems: Non-Livingston poems A Widow Lady 46.154 Aleon 50.000 Almeria 33.333 Ancient Poetry 30.000 Autumn Elegy 13.636 Banning Ramsay 45.455 Belle's Progress 38.462 Broken Flute 33.333 Brown Cow 47.826 Cleora 23.810 Columbia 43.750 Cornelia Remsen 59.259 Death of Hillard 30.000 Delia Crusca 46.667 Dutch People 54.545 Epithalamium 42.857 Fox and Cat 50.000 Friendship 39.130 Full Blown Rose 41.026 Indian Eclogue 46.154 Juliet 31.250 Laura 45.455 Logan's Triumphs 50.000 Lover's Vows 15.000 Major Wyllys 48.485 Mercy 61.538 Miniature Profile 36.364 Miranda's Birthday 36.667 Miss Anna D-nd-s 45.946 Mr. Brakenridge 50.000 Ode to Learning 41.176 Philadelphia 66.667 Present Hour 53.333 Rich and Ambitious 46.667 Sailor Boy 64.706 Serenade 33.333 Situation in France 44.444 Sylph 54.054 To Cymon 57.143 To Florio 40.000 Vision of Franklin 20.000 Wedding Ring 78.571 Totals: Percent L-pairs 43.387 Mean percentage for 43.719 individual poems: Night Before Christmas Percent L-pairs 77.551
MacDonald P. Jackson
UNIVERSITY OF AUCKLAND, NEW ZEALAND
I am grateful to Mary Van Deusen for compiling data for analysis, computer programmer Paul Kosinski for creating lists of frequencies, and linguist Lyn Bates for help with transcriptions into Arpabet.
MACDONALD P. JACKSON is professor emeritus of English at the University of Aucldand and a fellow of the Royal Society of New Zealand. He was an associate general editor of, and contributor to, the award-winning Oxford Thomas Middleton: The Collected Works (2007) and its companion volume Thomas Middleton and Early Modern Textual Culture (2007). He has coedited three volumes of Cambridge's The Works of John Webster (2003, 2007, in press, 2017). Among his most recent publications are Who Wrote "The Night Before Christmas"? (2016) and three essays in The New Oxford Shakespeare Authorship Companion (2017).
Anon. "Epitaph of Reuben Chase." 2016, http://www.webnests.eom/Chase/chronicles/epitaphreuben.htm. Accessed 22 Feb. 2017.
--. "Reuben Chase." 2016, https://en.wikipedia.org/wiki/Reuben_Chase. Accessed 22 Feb. 2017.
"Arpabet." Wikipedia. 2017, https://en.wikipedia.org/wiki/Arpabet. Accessed 22 Feb. 2017.
Burrows, Edwin G., and Mike Wallace. Gotham: A History of New York City to 1898. Oxford UP, 1999.
Burrows, John. "'Delta': A Measure of Stylistic Difference and a Guide to Likely Authorship." Literary and Linguistic Computing, vol. 17, no. 3, 2002, pp. 267-86.
The CMU Pronouncing Dictionary. 2017, http://www.speech.cs.cmu.edu/cgi-bin/cmudict. Accessed 22 Feb. 2017.
Foster, Don. Author Unknown: On the Trail of Anonymous. Henry Holt, 2001.
Jackson, MacDonald P. Who Wrote 'The Night Before Christmas'?: Analyzing the Clement Clarke Moore vs. Henry Livingston Question. McFarland, 2016.
--. "Response by MacDonald P. Jackson." 2017, melvilliana.blogspot.co.nz/2017/02/ response-by-macdonald-p-jackson.html/. Accessed 22 Feb. 2017.
Juola, Patrick. "Authorship Attribution." Foundations and Trends in Information Retrieval, vol. 1, no. 3, 2008, pp. 233-334.
Kaller, Seth. "The Authorship of The Night Before Christmas." 2004, https//www.sethkaller.com/about/education/tnbc/. Accessed 22 Feb. 2017.
Moore, Clement Clarke. Poems. Bartlett and Welford, 1844.
Nickell, Joe. "The Case of the Christmas Poem, Part 1 and Part 2." Manuscript, vol. 54, no. 4, 2002, pp. 293-308, and vol. 55, no.i, 2003, pp. 5-15.
Nissenbaum, Stephen. "There Arose Such a Clatter: Who Really Wrote 'The Night Before Christmas'? (And Why Does It Matter?)". Common-Place, vol. 1, no. 2, 2001, http:// www.common-place.org/vol-01/no-02/moore/.
Norsworthy, Scott. "Clement C. Moore's Published Letter on his Authorship of 'Visit from St. Nicholas.'" 2017, melvilliana.blogsport.co.nz/2017/01/clement-c-moores-published-letter-on.html/. Accessed 22 Feb. 2017.
Caption: Figure 1 * Distribution of Livingston and Non-Livingston poems having percentages of Livingston-favored common words falling within various ranges. Position of "The Night Before Christmas" also shown. Relevant figures are listed in Table 1.
Caption: Figure 2 * Distribution of Livingston and Non-Livingston poems having percentages of Livingston-favored phoneme pairs falling within various ranges. Position of "The Night Before Christmas" also shown. Relevant figures are listed in Table 2.
|Printer friendly Cite/link Email Feedback|
|Author:||Jackson, MacDonald P.|
|Article Type:||Critical essay|
|Date:||Dec 22, 2017|
|Previous Article:||"Softened" Voice Quality in Poetry Reading: An Acoustic Study.|
|Next Article:||How to Build a Metaphor: Novel Metaphors Construed by Concrete Elements in Tomas Transtromer's Poetry.|