Hostname: page-component-76fb5796d-2lccl Total loading time: 0 Render date: 2024-04-26T01:12:14.534Z Has data issue: false hasContentIssue false

Hobbes and “A Discourse of Laws”: The Perils of Wordprint Analysis

Published online by Cambridge University Press:  05 August 2009

Extract

John Hilton, Noel Reynolds, and Arlene Saxonhouse have employed statistical authorship analysis to claim that Hobbes authored three anonymously published works. But the authors of the study were not aware of close textual parallels between one discourse and a recently discovered work of Francis Bacon which argue against this attribution and call attention to deficiencies in the method of proof. Compounding this difficulty, they could not determine the stylistic pattern of Bacon, who is a candidate for authorship of the three discourses. It can be concluded that the authorship of at least one of the discourses is still a mystery, and that while the science of statistical wordprint analysis has advanced significantly over the past thirty years, it still has limitations which must be taken into account in any authorship study.

Type
Research Article
Copyright
Copyright © University of Notre Dame 1997

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 Hobbes published his translation of Thucydides (1627) at age 39. The major works of Hobbes's own hand were published much later: Elements of Law (1640) at age 52; De Cive (1642) at age 54; Leviathan (1651) at age 63.

2 Reynolds, Noel B. and Hilton, John L., “Statistical Wordprint Analysis Identifies New Hobbes Essays”, International Hobbes Association Newsletter 14 (1992)Google Scholar; Thomas Hobbes and the Authorship of the Horae Subsecivae”, History of Political Thought 14 (1994)Google Scholar. Reynolds and Saxonhouse have published the work under Hobbes's name with several explanatory essays: Hobbes, Thomas, Three Discourses, ed. Reynolds, Noel B. and Saxonhouse, Arlene W. (Chicago: University of Chicago Press, 1995).Google Scholar

3 Horae Subsecivae: Observations and Discourses (London: Edward Blount, 1620)Google Scholar, hereafter referred to as Horae. “Horae subsecivae” can be translated as “leisure hours”. Reynolds, and Saxonhouse, , “Hobbes and the Horae Subsecivae” in Three Discourses, p. 3 n. 1.Google Scholar

4 Reynolds, and Hilton, , “Statistical Wordprint”, p. 5.Google Scholar

5 Reynolds, and Saxonhouse, , “Hobbes and Horae”, p. 16.Google Scholar

6 Honan, William H., “3 Hobbes Essays Renew Debate over Machiavelli,” New York Times, 20 12 1995, p. B14.Google Scholar

7 Grey Brydges and Gilbert Cavendish were once considered candidates, but subsequent evidence has ruled them out of consideration. Reynolds, and Saxonhouse, , “Hobbes and Horae,” p. 9.Google Scholar

8 Bickley, Francis, Cavendish Family (Boston and New York: Houghton & Miflin, 1914), pp. 1135.Google Scholar

9 Reik, Miriam, The Golden Lands of Thomas Hobbes (Detroit: Wayne State University Press, 1977), p. 30Google Scholar. Aubrey reports that the principal of Magdalen Hall who recommended Hobbes was named Sir Hussey, James. John Aubrey: Brief Lives, ed. Barber, Richard (Rochester: Boydell Press, 1982), p. 151.Google Scholar

10 Reik, , Golden Lands, p. 30Google Scholar; Rogow, Arnold, Thomas Hobbes: Radical in the Service of Reaction (New York: W. W. Norton, 1986), pp. 5764Google Scholar; Saxonhouse, , “Hobbes and the Horae Subsecivae,” Polity 13 (1981): 544–48.CrossRefGoogle Scholar

11 Records of the Virginia Company of London, ed. Kingsbury, S. M. (Washington, 1906), 2: 40Google Scholar cited in Malcolm, Noel, “Hobbes, Sandys, and the Virginia Company”, The Historical Journal 24 (1981): 298.CrossRefGoogle Scholar

12 Thucydides, , The Peloponnesian War, trans. Hobbes, Thomas (Chicago: University of Chicago Press, 1989), pp. xix–xx.CrossRefGoogle Scholar

13 Thomae Hobbes Malmesburiensis Opera (Aalen: Scientia, 1961), 5: 323–40.Google Scholar

14 Bickley, , Cavendish Family, p. 45.Google Scholar

15 Tuck, Richard, Philosophy and Government, 1572–1651 (Cambridge and New York: Cambridge University Press, 1993), pp. 283–84.CrossRefGoogle Scholar

16 Rogow, , Thomas Hobbes, p. 60Google Scholar; Robertson, George Croom, Hobbes (1910; reprint, St. Clair Shores: Scholarly Press, 1970), p. 203.Google Scholar

17 Bickley, , Cavendish Family, pp. 5960.Google Scholar

18 Spedding, , and Heath, , eds., Works of Francis Bacon, (London: Longmans, Green, Reader and Dyer, 1874), 14: 542.Google Scholar

19 Malcolm, , “Hobbes, Sandys,” p. 316.Google Scholar

20 Robertson, , Hobbes, p. 19 n. 2.Google Scholar

21 Bacon, , Works, 10: 60.Google Scholar

22 Aubrey calls the work “Of Cities,” but he almost certainly means “True Greatness of Kingdoms and Estates” which was translated into Latin and incorporated into the De Augmentis. Brief Lives, p. 151.

23 “il seroit bon que je sceusse precisement ce qu'il me semble qu'on m'assura des-long-tems Que vous avez este secretaire des Etudes de ce Chancelier Bacon” (The Correspondence of Thomas Hobbes, ed. Malcolm, Noel [Oxford: Oxford University Press, 1994], 2: 624, 627–28)Google Scholar. Verdus (1621–1675) was an admirer of Hobbes and translated Bacon's Wisdom of the Ancients and Hobbes's De Corpore, De Homine, De Cive, and Leviathan into French. Ibid., 1: li–lii; 2: 904–14.

24 Peck, Linda Levy, “Hobbes on the Grand Tour: Paris, Venice, or LondonJournal of the History of Ideas 57 (1996): 177–82CrossRefGoogle Scholar. Malcolm, Noel, De Dominis (London: Strickland and Scott, 1984), p. 120 n. 280Google Scholar

25 For an excellent treatment of Sarpi's life and thought, see Wootton, David, Paolo Sarpi: Between Renaissance and Enlightenment (Cambridge: Cambridge University Press, 1983).CrossRefGoogle Scholar

26 Vita del Padre Paolo, dell'Ordie de' Servi, e teolog della serenissima republ di Venetia (Leida, 1646)Google Scholar. For questions about Micanzio's dedication to the church, see Wootton, , Paolo Sarpi, pp. 101–03, 109, 138.Google Scholar

27 The inscription on one of the manuscripts reads “Translated out of the original Italian letters by Th: Hobbes secretary to Lord Cavendish” (Gabrieli, Vittorio, “Bacone, La Riforma e Roma Nella Versione Hobbesiana d'un Carteggio di Fulgencio Micanzio,” English Miscellany 8, pp. 232–33 fig. 5).Google Scholar

28 Most critics agree that the letters were translated for Cavendish's benefit, who while he spoke Italian, may not have had an expert command of the language. Malcolm argues for his knowledge of Italian citing his introduction of a Venetian ambassador, while Tuck implies his lack of understanding. De Dominis, pp. 49–50; Hobbes (Oxford and New York: Oxford University Press, 1989), p. 5.Google Scholar

29 Rahe, Paul, Republics Ancient and Modern: Classical Republicanism and the American Revolution (Chapel Hill: University of North Carolina Press, 1992), p. 367Google Scholar. Rahe's suggestion that the translation was circulated is bolstered by the fact that there are two copies, one at Chatsworth and the other at the British Library. If Hobbes were translating for Cavendish alone, no second copy would have been needed.

30 Rawley, Bacon's chaplain, published some of Bacon's works in the “Resuscitatio” in 1657, and quoted from one of Micanzio's letters to Cavendish. Works, 1: 15.Google Scholar

31 Micanzio to Cavendish (17 June 1616), Gabrieli, , “Bacone,” p. 205.Google Scholar

32 “I am exceedingly bound to you for relating to Sir Francis Bacon how much I esteem his judgements and learning, having not a long time met with any writing that has given me greater content and having taken such a conceit of the Author from his Essays that I find myself very much carried away to love and honour him. But then the favour of calling me to a commerce of letters with him I esteem for singular and accompt in a great happiness and a great debt of mine to you. I write to him in Italian, supposing he understands the tongue, for Latin I write with difficulty as ordinary Italians do. But if need be I will write in Latin, not caring though I do it ill so I may enjoy the reciprocal communication of so lovely a Person whom I image to myself of a temperature every way perfect“ (Micanzio to Cavendish [31 March 1616], Ibid., pp. 203–204).

33 Bacon, , “Epistolae ad fulgentium,” Works, 14: 530–32.Google Scholar

34 Rahe, , Republics, p. 367Google Scholar, citing Gabrieli.

35 Malcolm, , “Hobbes, Sandys,” p. 316Google Scholar citing Gabrieli p. 203; Rahe, , Republics, p. 367Google Scholar; Tuck, , Hobbes, p. 10Google Scholar; Reynolds, and Hilton, “Hobbes and Authorship”, p. 363Google Scholar citing correspondence with Malcolm.

36 Strauss, Leo, letter to Kojeve, 19 04 1934 in On Tyranny, (New York: Free Press, 1991), p. 225Google Scholar. See also Strauss, , Political Philosophy of Hobbes, p. xii n. 1Google Scholar. See also Wolf, F. O., Die neue Wissenschaft des Thomas Hobbes: Zu den Grundlagen der Politischen Philosophie der Neuzeit. Mit Hobbes' Essays (Stuttgart-Bad Cannstatt: Friedrich Frommann Verlag, 1969), pp. 116,133–34.Google Scholar

37 Saxonhouse, , “Hobbes and Horae,” pp. 544–45.Google ScholarBush, DouglasHorae Subsecivae” in Times Literary Supplement (31 06 1943), p. 367Google Scholar. Wordprint analysis did not find the “Discourse on Flatterie” to be consistent with Hobbes's writing style. Reynolds, and Saxonhouse, “Hobbes and Horae,” p. 19.Google Scholar

38 Reynolds, and Saxonhouse, , “Hobbes and Horae,” p. 8Google Scholar

39 Ibid., p. 4 n. 2.

40 Saxonhouse, , “Hobbes and Horae,” pp. 546–48.Google Scholar

41 Ibid., p. 547. For the opposite view, see Malcolm, “Hobbes, Sandys”: “if it had been the work of an insignificant secretary, its seem[s] impossible to imagine why it should not have been published under the name of its author” (p. 321).

42 Saxonhouse, , “Hobbes and Horae,” pp. 546–48Google Scholar

43 O'Brien, Edward “‘Horae Subsecivae ’ 1620”, Notes and Queries, 12: 101103, 162–63Google Scholar. O'Brien makes too sweeping a claim, namely, that only Bacon and the author of the Horae wrote essays in this period. Matthews shows that there were other contemporaries who wrote essays, but it is not an exaggeration to say that Bacon introduced the essay to England. Matthews, C. Elkin, “Commentary on O'Brian,” Notes and Queries, 12: 164Google Scholar. See also Faulkner's, Robert forthcoming piece on the Baconian essay in Encyclopedia of the Essay, ed. Chevalier, Tracy (Fitzroy-Dearborn: London, nd).Google Scholar

44 Even Strauss, who champions Hobbes as the author of the “Essayes,” admits there are “considerable borrowings” from Bacon's essays. Strauss, , Political Philosophy of Hobbes, p. xii n 1Google Scholar. Saxonhouse and Reynolds believe that the essays of the Horae are “reminiscent in style and language of Bacon's essays” (“Hobbes and Horae,” p. 4).

45 The essay in the Horae is entitled “Of Expences ”.

46 Saxonhouse, “Hobbes and Horae, p. 550.Google Scholar

47 Ibid., p. 550; Faulkner, forthcoming piece on the Baconian essay.

48 Reynolds, , “Statistical Wordprinting,” in Three Discourses, p. 157.Google Scholar

49 For a list of all sixty-five of Morton's noncontextual-word patterns, see Three Discourses, Appendix 3, pp. 166–69.

50 Hilton, John L., “On Verifying Wordprint Studies: Book of Mormon Authorship,” BYU Studies 30 (1990): 91.Google Scholar

51 The known works of Hobbes tested were “On the Life and History of Thucydides”, Leviathan, Elements of Law, and Behemoth. See Three Discourses, Appendix 2, p. 165.

52 Reynolds, and Saxonhouse, , “Hobbes and Horae,” p. 12.Google Scholar

53 Ibid., p. 12.

54 “For each of the 65 word patterns compared, the null hypothesis is that there will be no difference between the texts larger than would be expected from random normal variation” (Reynolds and Saxonhouse, Ibid., p. 15 n. 21).

55 Reynolds, , “Statistical Wordprinting,” pp. 160–61.Google Scholar

56 Reynolds, and Saxonhouse, , “Hobbes and Horae,” p. 15.Google Scholar

57 Ibid., pp. 14–15.

58 Ibid., pp. 9–10.

59 The Latin text and an English translation can be found in Neustadt, Mark, “The Making of the Instauration: Science, Politics and Law in the Career of Francis Bacon” (Ph.D. diss., Johns Hopkins University, 1987)Google Scholar, hereafter referred to as “Aphorisms.”

60 The manuscript originated in Hardwick house, the other Cavendish house, but the libraries were combined in 1815. Catalogue of the Library ofChatsworth, (London: Chiswick Press, 1879), p. xvii.Google Scholar

61 Neustadt, , “Making of the Installation,” p. 241.Google Scholar

62 Rogow, , Thomas Hobbes: “As William's tutor, he had access to Hardwick's library—which Hobbes considerably expanded beyond the total of six books” (p. 62)Google Scholar. Hobbes also catalogued Chatsworth library in the late 1630s. Saxonhouse, , “Hobbes and Horae,” pp. 547–48.Google Scholar

63 Neustadt, , “Making of the Instauration,” p. 242.Google Scholar

64 Ibid., p. 243.

65 Neustadt notes two parallels between the Maxims of Law and the “Aphorisms,” Ibid., p. 282 n. 7 and p. 297 n. 14. Aphorism 13 discusses attempts to prevent changing laws in the future, which is elaborated in Regula XIX, Works, 7:370–72Google Scholar. Aphorism 19 explains the relationship between public right and private right, closely paralleling Regula V, Works, 7: 344–45Google Scholar. In addition, Bacon discusses the inability of a supreme power to be bound in the future in History of the Reign of King Henry VII, ed. Weinberger, Jerr (Ithaca and London: Cornell University Press, 1996), pp. 135–36Google Scholar and in Maxims, Regula XIX, Works, 7: 369–72Google Scholar. These treatments have a parallel in “Aphorisms,” [13–14].

66 For other acccounts of the “Aphorisms,” see Coquillette, Daniel R., Francis Bacon (Stanford: Stanford University Press, 1992), pp. 237–44Google Scholar; Faulkner, Robert K., Francis Bacon and the Project of Progress (Lanham, MD: Rowman and Littlefield, 1993), pp. 215–19Google Scholar; Neustadt, , “Making of the Installation,” pp. 138–43.Google Scholar

67 Saxonhouse, , “Hobbes and Modern Political Thought,” in Three Discourses, p. 150.Google Scholar

68 E.g., in listing seven benefits of proper execution of laws, the discourse echoes themes from Bacon's “Of Judicature.” The proper execution of the application of the law by a judge must: (1) reassure parties that the law is equally enforced, (2)suppress force and fraud, (3) protect private property, (4) ensure that the rich do not dominate the poor.

69 Reynolds, and Hilton, , “Thomas Hobbes,” p. 367.Google Scholar

70 Reynolds, , “Statistical Wordprinting,” pp. 161–62.Google Scholar

71 Reynolds, and Hilton, “Thomas Hobbes,” p. 368.Google Scholar

72 “While there are many unanswered questions relating to wordprint analysis as applied to translations, we might expect that an author's own translation of his works from his academic language to his native tongue might show up in wordprint analysis as his own work” (Ibid., p. 368). While the authors are hopeful and “expect” that an author's style of writing in English might be preserved when translating his own text from a foreign language, such an expectation is neither supported nor denied by evidence. See Hilton, , “Mormon Wordprint,” p. 97.Google Scholar

73 Reynolds, and Hilton, “Thomas Hobbes,” p. 369.Google Scholar

74 Reynolds, and Saxonhouse, , “Hobbes and Horae,” p. 13, n. 19.Google Scholar

75 See Montagu, Basil, Works of Francis Bacon (Parry and McMillan: Philadelphia, 1855), 3: 329Google Scholar. There were earlier English translations of portions of the Novum Organum, but none in Bacon's lifetime. In 1640, Watt included a translation of the introduction to theNovum Organum along with his translation of the De Augmentis. In 1671, three short portions of the New Organon appeared in the Resuscatio. In 1671, there was an abridged translation of the New Organon in the tenth edition of Sylva Sylvarum.

76 Bacon, Francis, Novum Organum, trans, and ed. Urbach, Peter and Gibson, John (Chicago: Open Court Press, 1994).Google Scholar

77 Reynolds stresses that the test is conservative, in that it “will not allow clear determinations of authorship in all cases but predicts ambiguity in about 30 percent of all comparisons.” This conservatism in the testing is seen in several examples: the test does not assume a normal distribution, it controls noise, it uses control authors, it samples testing blocks several times from different starting points. Apparently these precautions are not enough to eliminate false positives. “Statistical Wordprinting,” pp. 158–61.

78 Reynolds, and Hilton, , “Thomas Hobbes,” p. 367Google Scholar

79 Reynolds, and Saxonhouse, , “Hobbes and Horae,” p. 11 n. 15, p. 12.Google Scholar

80 The first block contains the original ten essays, first published in 1597, although the authors used the 1626, expanded versions of these essays. The second contains twelve essays added in either the 1612 or 1625 editions. The authors recognize that the essays are from several periods and hope to address it in the future. Reynolds, and Hilton, , “Thomas Hobbes,” p. 368.Google Scholar

81 Bacon revised and expanded other works as well. For example, the New Organon, used in the Bacon sample, contains parts of works composed earlier. The De Augmentis contains several Latin translations of Bacon's previously published essays. See Urbach, and Gibson, , “Editor's Introduction,” Novum Organum, pp. xi–xiiGoogle Scholar. See also Coquillette, , Francis Bacon, p. 226Google Scholar. Reynolds admits that heavily edited texts might not be suitable for testing (“Statistical Wordprinting,” p. 161).

82 Hilton, , “Book of Mormon,” pp. 9293Google Scholar. Mosteller, Frederick and Wallace, David L., Applied Bayesian and Classical Inference: The Case of the Federalist Papers (New York: Springer-Verlag, 1984).CrossRefGoogle Scholar