Neville’s surviving letters show him reporting uncertain news, bargaining over trade and defending choices under pressure. His request to Cecil for time to settle his estate before the French embassy is one example of prose doing practical work. The collection reaches us unevenly, through manuscripts, copies, transcriptions and letters printed in Edmund Sawyer’s edition of Winwood’s papers; together they allow his language to be followed across years and correspondents.
The research collection counted on 20 September 2026 contained 169 letter records and three longer documents. Excluding headings, addresses, enclosed and separately marked quoted material, and an identified duplicate letter, it contains approximately 136,800 words: about 129,600 in the letters and 7,200 in Neville’s Case, his confession and his Advice.12
This chapter describes the collection as counted on 20 September 2026; Chapter 12 now reports the bigram rerun of 3 October, after removal of the duplicate letter record. The figures reflect different dates, selections and counting rules:
| Count | Descriptive census here | Bigram study in Chapter 12 |
|---|---|---|
| Records | 169 letters + 3 documents | 165 English letters + 3 documents |
| Words counted | 136,764 normalized word tokens after the stated exclusions | 137,551 alphabetic lemma tokens under the study’s selection rules |
| Parallel witnesses | Both listed; the duplicate letter counted once in the word total | Duplicate removed; one witness retained |
The bigram selection excludes three French records but retains enclosed and quoted material. It counts word tokens whose base forms contain letters, omitting purely numerical entries; the earlier census counts normalized word tokens under its stated exclusions. Both word totals count the identified duplicate letter only once.
The shape of the surviving letters
The collection is richest during Neville’s French embassy. The September census’s date labels place 56 letter records in 1599 and 40 in 1600: 96 records, or 56.8 per cent of the letters. Twenty more fall in 1601. The earlier years are sparsely represented, and the later correspondence survives in smaller groups.
| Year | Records | Year | Records | Year | Records |
|---|---|---|---|---|---|
| 1590 | 1 | 1598 | 2 | 1606 | 5 |
| 1591 | 0 | 1599 | 56 | 1607 | 2 |
| 1592 | 0 | 1600 | 40 | 1608 | 7 |
| 1593 | 1 | 1601 | 20 | 1609 | 3 |
| 1594 | 2 | 1602 | 8 | 1610 | 0 |
| 1595 | 1 | 1603 | 1 | 1611 | 0 |
| 1596 | 0 | 1604 | 4 | 1612 | 8 |
| 1597 | 0 | 1605 | 4 | 1613 | 4 |
Counts follow the September collection’s date labels. Zero means no record in this collection, not that Neville wrote no letters that year.
The concentration defines the comparison: an apparent habit of his language may belong particularly to embassy dispatches. Later political counsel and personal letters provide useful tests of whether that habit persists elsewhere.
The recipients show a similar concentration. Robert Cecil accounts for 106 records, Ralph Winwood for thirty and Thomas Windebank for eight. Together they receive 85.2 per cent of the letters. The other twenty-five records range across sovereigns, officials, friends and local correspondents. These figures describe the assembled collection. They cannot rank the importance of relationships whose correspondence has been lost.
What each witness preserves
The texts reach us through several kinds of source.
Sawyer’s edition appeared in 1725. Its spelling, punctuation and typography cannot establish how Neville wrote the original manuscript. The book therefore modernizes quotations from it while preserving vocabulary, syntax and meaning. A spelling claim requires the manuscript or a suitable diplomatic transcription. A handwriting identification requires examination of the manuscript image and supporting attribution evidence.
John O’Donnell’s transcriptions provide the principal human-transcribed body of letters. Such a transcription can preserve forms lost in later print, including abbreviations and original spellings. Its value depends on the particular source and the work already done on it: identifying the transcriber, locating an image and checking a difficult passage are different stages.
Three longer works
The letters are accompanied by three documents that widen the view of Neville’s prose. Their occasions differ, and those differences are worth preserving.
His Case, printed by Sawyer from a paper the editor described as entirely in Neville’s own hand, reconstructs his involvement in the events surrounding the Essex rising. It gives an extended account of intention, meetings and decisions under the pressure of accusation. The separate declaration of 2 March 1601, usually called his confession, survives in the original manuscript in Neville’s hand. The two documents allow us to examine how he explains conduct and judges what he had understood, intended or failed to do. They are two separate witnesses about the same crisis in Neville’s life.3
Neville’s Advice turns from personal defence to political counsel. It considers the king’s financial difficulties and the conditions under which Parliament might offer relief. Here the writer weighs courses of action and their consequences for the relationship between king and subjects. Gardiner’s printed edition provides a public text for comparison; the surviving manuscript copies require their own collation. Neville acknowledged writing the paper in the Commons on 14 May 1614. For the Advice, this book follows Gardiner’s edition and the manuscript witnesses rather than the corpus transcription. Authorship and the readings of particular copies are separate questions.4
Checking the text beneath the count
Searches use a consistent text so spelling variation does not hide a phrase. Where an identified source supplies a different original form, the collection preserves it separately. Spelling comparisons must use the appropriate source forms.1
From surviving prose to comparison
The collection permits tests of a word in dated correspondence, a recurring sequence of argument, or language across recipients. Each can be rerun with the documents removed or the printed and manuscript-transcribed groups examined separately.
A result dominated by Cecil dispatches requires a different explanation from one found across private letters, diplomatic reporting and political counsel.
Chapter 12 compares recurring two-word sequences across the plays. Chapter 13 tests a set of linguistic features selected from correspondence against 236 plays. Chapter 14 asks whether diplomatic vocabulary selected without Shakespeare counts changes at the predicted time. The figures in each study refer to the corpus version and selection specified there, which may differ from the September census given here.
Notes and sources
Counts were recalculated from the author’s current research collection on 20 September 2026: 169 letter records and 3 document records. The word totals are 129,597 for the letters and 7,167 for the documents under the stated exclusions. They count the collection’s normalized words, retaining salutations, subscriptions and reported speech; separately marked quotations are excluded. These describe the encoded collection, including its declared parallel witnesses and unresolved records, rather than a census of all surviving original letters. The principal printed source is Edmund Sawyer, ed., Memorials of Affairs of State in the Reigns of Queen Elizabeth and King James I, 3 vols. (London, 1725): volume I, volume II, volume III. The counts follow the stated source classifications; they do not certify every record’s text or date. The separate 1884 witness is W. D. Macray, ed., Beaumont Papers (London: Nichols and Sons, 1884), pp.13–14, which supplies “[1606?]” as an editorial date and “Wednesday night” in the dateline. For the enclosed royal response subscribed De Neufville, Sawyer I, pp.87–88. That reply is excluded from the word totals, as are the separately enclosed memorials. The year and recipient tables count letter records, including the parallel witness; their date labels are not all equally certain. There are 4,289 attributes preserving original word forms and twenty-three continuation markers, used when normalization splits a source form across tokens. These are features of the edition’s encoding, not additional words written by Neville. A spelling study must use the appropriate source forms and avoid counting a continuation twice.↩︎
The 4 June 1606 letter to Winwood survives in two records, one from Sawyer and one transcribed by John O’Donnell. Counting the identified pair once leaves 168 communications, subject to any further duplicate identifications; the word total uses O’Donnell’s witness. See Sawyer, Memorials, II, pp. 216–217. The printed 4 June date corresponds to 14 June in the Gregorian calendar.↩︎
“Sir Henry Nevill’s Case,” Sawyer, Memorials, I, pp. 302–304; Neville’s declaration, 2 March 1601, The National Archives, SP 12/279, f. 15, calendared in Calendar of State Papers, Domestic, Elizabeth, 1601–1603, pp. 5–8, no. 11. The confession’s original is in Neville’s handwriting; the text follows John O’Donnell’s transcription. Full-document collation is a distinct task. The initial repair of the Case ending and five shorter readings added 202 encoded words at that stage; subsequent source and tokenization changes mean this is not the net change underlying the current whole-collection total. The precise composition date of the Case is unestablished: Sawyer’s heading supplies no exact date. The analytical record uses 1601 as an inferred contextual year.↩︎
Neville, “An Advice Touching the Holding of a Parliament,” in Samuel Rawson Gardiner, History of England from the Accession of James I to the Disgrace of Chief-Justice Coke, II (London, 1863), Appendix V, pp. 389–394. For Neville’s acknowledgment in the Commons, see Clayton Roberts and Owen Duncan, “The Parliamentary Undertaking of 1614,” English Historical Review 93, no. 368 (1978), pp. 481–498, especially p. 493, article. The book’s Chapter 9 gives the fuller political context and source comparison.↩︎