# Punctuation and syntax habits (rates per 1000 words unless stated)

Controls: 115 samples. z = (value - control mean) / control sd. Beale texts: letters (3 letters, 2716 words), narrative (3954 words), Morriss statement (865 words).

| feature | letters | narrative | Morriss | z letters | z narrative | controls mean (sd) | early 1805-46 mean | late 1858-1900 mean |
|---|---|---|---|---|---|---|---|---|
| semicolons | 1.5 | 7.1 | 7.0 | -0.9 | -0.1 | 7.9 (7.4) | 7.2 | 8.7 |
| colons | 1.1 | 2.0 | 0.0 | -0.3 | +0.2 | 1.6 (1.8) | 1.6 | 1.6 |
| dashes | 2.2 | 0.8 | 0.0 | -0.3 | -0.6 | 4.0 (5.5) | 2.2 | 5.5 |
| parentheses | 0.0 | 0.0 | 0.0 | -0.7 | -0.7 | 1.0 (1.3) | 0.8 | 1.1 |
| commas | 97.9 | 95.0 | 102.3 | +0.4 | +0.3 | 85.5 (29.9) | 85.5 | 85.5 |
| exclamations | 0.0 | 0.0 | 0.0 | -0.4 | -0.4 | 1.2 (2.8) | 0.6 | 1.7 |
| questions | 0.0 | 0.0 | 0.0 | -0.6 | -0.6 | 1.4 (2.4) | 0.7 | 1.8 |
| etc_&c | 0.7 | 0.0 | 0.0 | +1.1 | -0.3 | 0.1 (0.5) | 0.2 | 0.1 |
| quotation_marks | 0.7 | 2.5 | 9.3 | -0.6 | -0.5 | 10.0 (16.1) | 3.9 | 15.2 |
| relativisers_which_who | 11.4 | 13.4 | 10.5 | +0.6 | +1.1 | 9.1 (4.0) | 10.6 | 7.6 |
| which | 8.1 | 8.6 | 5.8 | +0.8 | +1.0 | 5.6 (3.0) | 6.9 | 4.4 |
| that | 8.5 | 15.2 | 10.5 | -0.5 | +0.9 | 11.0 (4.9) | 11.0 | 11.0 |
| infinitives | 23.2 | 19.5 | 10.5 | +1.1 | +0.4 | 17.0 (5.6) | 16.6 | 17.3 |
| coordinators_and_but_or_nor | 48.2 | 47.9 | 66.3 | +0.5 | +0.5 | 44.3 (7.8) | 43.6 | 45.7 |
| and | 38.6 | 35.7 | 51.2 | +0.4 | +0.0 | 35.4 (7.5) | 34.3 | 37.3 |
| however | 1.1 | 2.0 | 1.2 | +0.9 | +2.4 | 0.6 (0.6) | 0.7 | 0.4 |
| doubtless | 1.1 | 0.3 | 0.0 | +5.3 | +1.0 | 0.1 (0.2) | 0.0 | 0.1 |
| upon | 2.9 | 3.3 | 2.3 | +0.7 | +0.9 | 1.8 (1.7) | 2.0 | 1.4 |
| until | 2.2 | 2.8 | 7.0 | +2.7 | +3.6 | 0.5 (0.6) | 0.5 | 0.5 |
| passive_be_+_ed | 18.0 | 13.9 | 14.0 | +1.9 | +0.9 | 10.0 (4.1) | 9.9 | 10.4 |
| negative_passive | 1.1 | 2.3 | 1.2 | +0.6 | +2.4 | 0.7 (0.7) | 0.6 | 0.8 |
| never_or_not_to_be_+_participle | 0.4 | 0.5 | 0.0 | +3.0 | +4.2 | 0.0 (0.1) | 0.0 | 0.0 |
| reflexive_pronouns | 4.8 | 3.0 | 3.5 | +2.5 | +1.0 | 1.8 (1.2) | 2.0 | 1.6 |
| untriggered_reflexives | 3.7 | 2.3 | 3.5 | +6.6 | +3.8 | 0.4 (0.5) | 0.4 | 0.3 |
| the_pct | 5.4 | 6.1 | 4.5 | -1.1 | -0.6 | 7.2 (1.7) | 7.4 | 7.1 |
| and_pct | 3.9 | 3.6 | 5.1 | +0.4 | +0.0 | 3.5 (0.7) | 3.4 | 3.7 |
| of_pct | 3.5 | 3.7 | 2.6 | -0.5 | -0.3 | 4.2 (1.4) | 4.5 | 3.8 |
| sentence_len_mean | 30.2 | 35.2 | 27.7 | +0.4 | +1.1 | 27.2 (7.2) | 28.9 | 25.9 |
| sentence_len_sd | 14.7 | 18.5 | 12.9 | -0.5 | +0.0 | 18.2 (7.4) | 18.7 | 17.8 |
| sentence_len_median | 28.0 | 32.0 | 22.0 | +0.7 | +1.3 | 23.4 (6.7) | 25.1 | 22.0 |
| long_sentences_40plus_pct | 23.3 | 28.6 | 12.9 | +0.3 | +0.8 | 19.9 (11.3) | 22.0 | 18.3 |
| paragraph_len_mean | 100.6 | 106.9 | 123.6 | -0.1 | -0.0 | 109.4 (73.5) | 124.6 | 97.0 |
| word_len_mean | 4.5 | 4.6 | 4.6 | -0.0 | +0.7 | 4.5 (0.2) | 4.5 | 4.5 |
| long_words_7plus_pct | 20.3 | 23.6 | 20.5 | -0.0 | +0.8 | 20.4 (4.2) | 20.5 | 20.5 |
| type_token_ratio_first1000 | 0.4 | 0.4 | 0.4 | -0.1 | +0.1 | 0.4 (0.0) | 0.4 | 0.4 |

Standardised Euclidean distance over these features (z-scored, RMS): calibration on control pairs.

Same-author pairs mean 1.100 (sd 0.453, n=203); different-author pairs mean 1.362 (sd 0.404, n=6352).

| pair | distance | %ile among diff-author pairs | %ile among same-author pairs | LR same:diff |
|---|---|---|---|---|
| beale_letters vs narrative | 1.160 | 33.5 | 69.5 | 0.76 |
| beale_letter1 vs narrative_A | 1.510 | 73.6 | 88.7 | 0.33 |
| beale_letters vs narrative_A | 1.200 | 38.8 | 73.4 | 0.67 |
| beale_letters vs narrative_B | 1.304 | 52.5 | 81.8 | 0.51 |
| narrative_A vs narrative_B | 0.954 | 10.5 | 43.3 | 1.61 |
| morriss vs narrative | 1.755 | 86.2 | 90.6 | 0.33 |
| morriss vs beale_letters | 1.911 | 90.1 | 92.6 | 0.52 |

Nearest control samples to beale_letters in feature space: jefferson_1 (1.52), jefferson_3 (1.52), lee_2 (1.56), cabell_2 (1.59), flint_3 (1.64), irving_3 (1.66)
Distance beale_letters to the other Beale text: 1.16

Nearest control samples to narrative in feature space: jefferson_3 (1.03), lee_2 (1.12), jackson_3 (1.15), ejames_2 (1.21), gregg_2 (1.26), irving_3 (1.31)
Distance narrative to the other Beale text: 1.16

Features where BOTH Beale texts deviate from the controls in the same direction (|z|>0.5): parentheses (-0.7/-0.7), questions (-0.6/-0.6), relativisers_which_who (+0.6/+1.1), which (+0.8/+1.0), however (+0.9/+2.4), doubtless (+5.3/+1.0), upon (+0.7/+0.9), until (+2.7/+3.6), passive_be_+_ed (+1.9/+0.9), negative_passive (+0.6/+2.4), never_or_not_to_be_+_participle (+3.0/+4.2), reflexive_pronouns (+2.5/+1.0), untriggered_reflexives (+6.6/+3.8), the_pct (-1.1/-0.6), sentence_len_median (+0.7/+1.3)

Features where they deviate in OPPOSITE directions (|z|>0.5): that (-0.5/+0.9)

| shared feature | letters | narrative | control samples >= min(letters,narr) | control authors with such a sample | which authors |
|---|---|---|---|---|---|
| parentheses | 0.0 | 0.0 | 50/115 | 22 | bagby, biddle, cabell, cable, christian, cooke, davis, eggleston, ejames, flint, gregg, hall, irving, jefferson, jqadams |
| questions | 0.0 | 0.0 | 54/115 | 22 | biddle, cabell, christian, cooke, davis, eggleston, ejames, flint, gregg, hall, irving, jefferson, jqadams, lee, page, p |
| relativisers_which_who | 11.4 | 13.4 | 30/115 | 15 | bagby, biddle, davis, eggleston, flint, gregg, hall, jackson, jefferson, jqadams, page, paulding, pike, poe, tjames |
| which | 8.1 | 8.6 | 22/115 | 15 | biddle, davis, eggleston, ejames, flint, gregg, hall, jackson, jefferson, jqadams, page, pattie, pike, poe, twain |
| however | 1.1 | 2.0 | 17/115 | 11 | biddle, cooke, gregg, irving, jackson, jefferson, lee, page, paulding, poe, shsp |
| doubtless | 1.1 | 0.3 | 12/115 | 6 | cabell, cooke, eggleston, flint, jefferson, pollock |
| upon | 2.9 | 3.3 | 24/115 | 12 | cable, cooke, davis, eggleston, ejames, flint, gregg, hall, jackson, jqadams, pattie, poe |
| until | 2.2 | 2.8 | 5/115 | 4 | christian, ejames, irving, randolph |
| passive_be_+_ed | 18.0 | 13.9 | 15/115 | 8 | cabell, christian, davis, gregg, hall, page, shsp, twain |
| negative_passive | 1.1 | 2.3 | 22/115 | 17 | cabell, cable, christian, cooke, davis, eggleston, ejames, flint, gregg, hall, jackson, jqadams, lee, page, pike, polloc |
| never_or_not_to_be_+_participle | 0.4 | 0.5 | 7/115 | 7 | ejames, jackson, jefferson, lee, page, paulding, pike |
| reflexive_pronouns | 4.8 | 3.0 | 13/115 | 11 | cabell, cable, eggleston, gregg, irving, jefferson, jqadams, lee, pattie, pike, twain |
| untriggered_reflexives | 3.7 | 2.3 | 1/115 | 1 | jefferson |
| the_pct | 5.4 | 6.1 | 25/115 | 12 | bagby, cabell, cable, jackson, lee, pattie, paulding, poe, pollock, randolph, tjames, twain |
| sentence_len_median | 28.0 | 32.0 | 31/115 | 14 | biddle, cabell, cooke, davis, eggleston, ejames, flint, gregg, hall, irving, jackson, jefferson, paulding, pollock |