# The Cave of Poverty, a poem. Written in imitation of Shakespeare. By Mr. Theobald

**Slug:** `cave_of_poverty_1715`  
**Priority:** 2  
**Author:** Lewis Theobald  
**Edition / date:** London: printed for Jonas Browne, sold by J. Roberts, 1715. FIRST (only) edition. Two copies obtained.

## Sources
- **archive.org** `bim_eighteenth-century_the-cave-of-poverty-a-p_theobald-mr-lewis_1715` - https://archive.org/details/bim_eighteenth-century_the-cave-of-poverty-a-p_theobald-mr-lewis_1715
  - catalogue title: The cave of poverty, a poem. Written in imitation of Shakespeare. By Mr. Theobald.  1715
  - date field: 1715; images/frames: 57; OCR engine: tesseract 5.3.0-3-g9920
  - digitised from / contributor: ['IA40313312-05', 'microfilm']
  - raw file(s) saved here (exactly as downloaded): `bim_eighteenth-century_the-cave-of-poverty-a-p_theobald-mr-lewis_1715_djvu.txt`
  - note: copy A: BL microfilm reel IA40313312-05; 57 frames
- **archive.org** `bib_fict_4103198` - https://archive.org/details/bib_fict_4103198
  - catalogue title: The cave of poverty, a poem. Written in imitation of Shakespeare. By Mr. Theobald
  - date field: 1715; images/frames: 70; OCR engine: tesseract 5.3.0-1-gd3a4
  - digitised from / contributor: The Johns Hopkins University Sheridan Libraries; publisher: London : Printed for Jonas Browne […], and sold by J[ames] Roberts
  - raw file(s) saved here (exactly as downloaded): `bib_fict_4103198_djvu.txt`
  - note: copy B: Johns Hopkins University (Sheridan Libraries) digitisation, 70 images incl. blanks; tesseract OCR

## Cleaned text(s)
- `cave_of_poverty_1715_copyA.clean.txt` - BL microfilm copy; 6,639 words; 93% of text lines pass a dictionary check
- `cave_of_poverty_1715_copyB_JHU.clean.txt` - JHU copy - fewer noise lines, better for the preface; 6,108 words; 90% of text lines pass a dictionary check

## Completeness
Complete (preface + poem in six-line stanzas). A third copy exists: bim_eighteenth-century_the-cave-of-poverty-a-p_theobald-mr-lewis_1715_0.

## OCR quality / notes
Cleaning: IA djvu.txt was normalised (ſ->s), microfilm target frames and junk lines trimmed, page numbers/signatures/running heads dropped, end-of-line hyphenation joined, and a CONSERVATIVE long-s correction applied (words with 'f' that are not in a modern frequency list are replaced by their 's' variant when that variant is common; a short override list covers so/she/sir/see/such/some...). Ambiguous real words (fame/same, fell/sell, fold/sold, found/sound, foul/soul, fit/sit ...) were left unchanged. Statistics of every change are in the `*.cleanstats.json` beside each cleaned file. Expect residual OCR noise; the '% of lines pass a dictionary check' figure above is a rough legibility score (modern-English word list; 18th-c. spellings count against it).