False Refinement and Declension:
Johnson on the History of the Language

Jack Lynch

Delivered at Samuel Johnson and the Languages of Literature: The First Annual Conference of the Johnson Centre, 12 September 1997, Birmingham, UK

"Every language," as we know, "has a time of rudeness antecedent to perfection, as well as of false refinement and declension."1 Now rise and fall together suggest a peak, and in this paper I'd like to discuss this linguistic trajectory by placing Johnson's comments on the history of the language in the context of other such statements from the fifteenth through the eighteenth centuries. My concern is the way these historical considerations work their way into discussions of linguistic propriety.

Johnson insists he has "studiously endeavoured to collect examples and authorities from the writers before the restoration, whose works I regard as the wells of English undefiled, as the pure sources of genuine diction." Purity of diction is by now a metaphor so thoroughly naturalized that it seems, if not dead, at least moribund. But it was not always so, and in calling Renaissance writers "undefiled" (a borrowing from Spenser) and "pure sources," he recalls a common habit of Renaissance humanists of castigating the undesirable as "impure." I'd like to suggest that these early humanists provide the best analogue to help us understand what Johnson had to say about the history of the English language.

Treating a language, whether classical or vernacular, as a substance capable of admitting adulteration was an idea developed for the first time in the fifteenth century, and was soon remarkably widespread. In antiquity and the Middle Ages there were ideas of good and bad Greek and Latin, but these almost never corresponded to pure and corrupt: such a dichotomy simply had not been conceived. Other divisions, of course, were common: language might be clear or muddled, decorous or inappropriate, belonging to the genus grande or genus humile. But it was not routinely conceived as a medium of "true" expression in which "false" elements served as pollutants. Only with the humanist "recovery" of "real" Latin in the fifteenth century (all these terms are, of course, hiding behind protective quotation marks) was the intervening millennium's language condemned as corrupt: scholastic philosophers and Barthollist lawyers were charged with infecting the Latin of the ancients with atrocious solecisms. In the place of these barbarians were erected monuments to the Augustan prose writers, especially Cicero, whose style became the hottest commodity in the literary marketplace. The urge was to cure the Latin tongue by excising its tumors, to restore it to the full bloom of health it experienced in the first century B.C. In the most extreme form of this project, radical Ciceronians would restrict the legitimate vocabulary to only those words that Cicero himself had used: Longolius in fact prided himself on using only the inflected forms found in Cicero's works. Such extremism prompted reactions even from the humanists' allies; Erasmus's Ciceronianus, for instance, is a satire on the absurd excesses of Longolius and his kind. But a less thoroughgoing variety of Ciceronianism was widespread throughout the Renaissance.

The whole enterprise depends upon the belief in a period in which a language most nearly approaches its true genius -- its peak -- and the humanists believed they found this in the years between the Jugurthine War and the death of Augustus. The sixteenth century therefore produced lexicons that drew illustrative quotations almost exclusively from antiquity -- not with the descriptive intentions that lie behind modern lexicographical corpora, but in the hope of regulating contemporary usage with that of the age of Cicero and Virgil.2 Terence's observation -- "nullumst iam dictum quod non dictum sit prius" -- became not a lament about the burden of the past, but a vow, a mustering of resolve, on proper style.3

Even those who rejected Ciceronianism itself continued to respect the ideal of linguistic purity, plotting a palace coup to dethrone Cicero from his position as stylistic arbiter, only to replace him with a junta consisting of Seneca, Lucan, Persius, and other writers of the Silver Age. The principle of grounding modern style on that of a privileged earlier age remained intact. Whether they looked to the writers of the Augustan or the Silver Age, Renaissance thinkers grounded their conception of style on a privileged moment in the history of the language.

The Ciceronian habit survived the radical stylistic and lexical experiments of the seventeenth century, and continued, albeit in altered form, into the eighteenth. We can detect the Renaissance concern with linguistic purity, for instance, in Johnson's declaration that he has "laboured to refine our language to grammatical purity, and to clear it from colloquial barbarisms, licentious idioms, and irregular combinations."4 Here we see the essentially Ciceronian arguments made not about Latin but about English. The movement was in fact already underway in the Restoration. Thomas Sprat hated the florid rhetorical habits of the Renaissance, and proposed a linguistic academy, calling for "a constant Resolution, to reject all the amplifications, digressions, and swellings of style: to return back to the primitive purity."5 Note that the hated amplifications and swellings are the legacy of the sixteenth-century Ciceronians: but note, too, that the expurgatory metaphor is also theirs. Sprat's choice of models is different, but his insistence on "return[ing] back to the primitive purity" demonstrates exactly the same dynamic. In revealing his anxiety about linguistic "infection," he uses the same old metaphors, just as the sixteenth-century anti-Ciceronians resorted to the conceptual categories of their rivals.

A similar call for an academy appears in the most famous eighteenth-century jeremiad on linguistic impurity, Swift's Proposal of 1712. Swift finds "our Language ... extremely imperfect; ... its daily Improvements are by no Means in Proportion to its daily Corruptions." Now, most eighteenth- century thinkers regarded the English Renaissance as a progress away from the rudeness and imperfections of the barbarous Middle Ages. But for many, the linguistic ascent from Chaucer and Lydgate to Shakespeare and Ben Jonson had its mirror image in a subsequent decline. Dryden, for example, writes, "That an Alteration is lately made in [our language] since the Writers of the last Age ... is manifest. ... But, that this is an Improvement of the Language, or an alteration for the better, will not so easily be granted. For many are of a contrary opinion, that the English tongue was then in the height of its perfection; that, from Jonsons time to ours, it has been in a continual declination." Swift has the same fear: "From that great Rebellion to this present Time, I am apt to doubt whether the Corruptions in our Language have not, at least, equalled the Refinements of it."6

Eighteenth-century writers, that is to say, saw contrary movements since the age of Elizabeth: progressive refinement toward Waller and Dryden on the one hand, yes, but a potentially unstoppable decline into babble on the other. Swift's contribution to Tatler 230 is well known: "I mean the continual Corruption of our English Tongue; which, without some timely Remedy, will suffer more by the false Refinements of Twenty Years past, than it hath been improved in the foregoing Hundred."7 Swift is in fact more cautious than this suggests: "the English Tongue," he says, "is not arrived to such a Degree of Perfection, as ... to make us apprehend any Thoughts of its Decay."8 But whether or not the decline had begun or was inevitable, anxiety about the threat was real, and caused the eighteenth century to inherit the humanistic idea of subjecting their own language to the discipline of a privileged earlier form, from a time before the decay began. Chesterfield therefore resolves, in the face of "All words, good and bad, ... jumbled indiscriminately together," that "The time for discrimination seems to be now come. Toleration, adoption and naturalization have run their lengths. Good order and authority are now necessary." Thus his advice to his son, in which he invokes the Terentian tag: "There is nothing truer than that old saying Nihil dictum quod non prius dictum."9

To judge by the front matter of the Dictionary alone, Johnson was attuned to this discourse of linguistic corruption. "Every language," remember, "has a time of rudeness antecedent to perfection, as well as of false refinement and declension." Languages follow the universal arc of ascent and decline: "Language proceeds," he says, "like every thing else, thro' improvement to degeneracy."10 Like Swift, he withholds comment on whether the English language had already in fact begun its decline. But the metaphors of the humanist tradition fill both the Preface and the History, where Johnson describes his task as "to guard the avenues of [his] languag[e], to retain fugitives, and repulse intruders."11 The intruders are those without a pedigree: only those words grounded in good, old usage are considered legitimate.

The question, then, for Johnson and his contemporaries, was which past to choose. Rome's golden age was easy to identify, and Italy and France had already discovered their own: as early as the sixteenth century, for instance, Bembo found Petrarch and Boccaccio models for imitation. Both of these names figure prominently, along with all greats of modern Italian literature, in the table of cited authors in the Accademia della Crusca's Vocabolario in 1612.12 In France, Richelet supported his Dictionnaire françois with "l'usage ... des bons auteurs de la langue françoise," most from the age of Louis Quatorze. English, on the other hand, was more troublesome. Even those who agreed that language had begun to decline differed on the time of the language's peak, and the many searches for a standard by which to regulate the language resulted in competing claims for various periods from the earliest times through the age of Anne.

Johnson hints at one common choice for a model of properly pure English when he identifies the modern-day barbarians in his Dictionary. "The whole fabrick and scheme of the English language," he writes, "is Gothick or Teutonick"; but "Our language ... has ... been gradually departing from its original Teutonick character, and deviating towards a Gallick structure and phraseology."13 The French, then, are the invaders, the poisoners of the pure Saxon streams. English is in a unique position among the European languages in having two parents, one Germanic, one Romance. By comparison with such relatively homogeneous languages as German and French, English therefore has an obvious standard of purity. The fit was indisputably imperfect, but the need for an authorizing corpus of pure diction was enough to put Old English, at least tentatively, in the position of Augustan Latin. Interest in Saxon antiquities, to be sure, was never really popular, but scholarly interest in the roots of the English language was considerable; and from at least the late sixteenth century this study was associated with British national identity. Against such a background, additions to the language from French sources could be seen as impurities. So Johnson calls souvenance "A French word which with many more is now happily disused." Gout ("taste") is "An affected cant word"; the second sense of to transpire is "a sense lately innovated from France, without necessity"; and ruse is "A French word neither eloquent nor necessary."

Whatever the attraction of the old Gothic vigor, though, it was a troublesome standard, for Teutonic words were routinely taken to task for their barbarism. Beowulf was not published until 1815, and no other work was available to play the role of an authorizing Saxon classic.14 Many in the early part of the eighteenth century therefore sought the language's acme in the decades immediately before them. The evidence here is lexicographical. Although Johnson was the first actually to produce an English dictionary with illustrative quotations, he was not the first to consider it, and his selection of many quotations from the sixteenth and early seventeenth centuries would not have been everyone's decision. Addison, for instance, planned a dictionary in which Tillotson, "the chief standard of our language," was to have provided the quotations.15 After asking "whether a word is English or not" and resolving to rely on "nothing but authority," Pope too looked to the late seventeenth century for his models: "Is [the word] in Sir William Temple, or Locke, or Tillotson? If it be, you may conclude that it is right."16 The list of authorities for his planned dictionary includes more names from after the Restoration than before it. But many of the things that made Old English a good choice made the late seventeenth century a bad one. Johnson's reference to "false refinement" sounds like the castigation of foppish and epicene French beaux popular since the Restoration, and catches the tone of many accusations against the seventeenth century. The affectation of a frenchified culture was simply not manly enough to satisfy the need for an English golden age.

So neither was a wholly adequate surrogate for Augustan Latin. Another option, though, was more suitable: the undefiled wells of "the writers before the restoration." In collecting his own quotations, Johnson turned to the Renaissance, when he felt the language reached its zenith: specifically, the 1580s through the 1650s, from Sidney to the Restoration, between rudeness and declension. His "studious" search for "examples and authorities from the writers before the restoration" is an expression of the eighteenth-century desire to authorize the language by reference to a corpus of canonical British texts. Johnson looked to the age of Elizabeth and the authors who, as Hawkins was the first to argue, provided the models for his own style.

He wasn't alone. Edward Phillips recommends "for the avoiding of ... gross words ... onely this in general, To be ever conversant in the best Authours, as Sir Philip Sidney, Sir Thomas More, Sir Walter Rawleigh, my Lord Verulam, Ben Johnson."17 For Swift, "the Period wherein the English Tongue received most Improvement" could be said "to commence with the Beginning of Queen Elizabeth's Reign, and to conclude with the great Rebellion in Forty-two." Even Hurd, whose temperament is often far from Johnson's, agrees with him on this point, finding "the condition of our language in the age of Elizabeth" to be "pure, strong, and perspicuous, without affectation."18

Tracing the ascent of this standard and justifying the primacy of the Renaissance is the business of Johnson's Preface and History. The "time of rudeness antecedent to perfection" took more than a millennium, from the fifth century to the end of the sixteenth. Indeed, in its earliest days, English had little to recommend it. "We know nothing of the scanty jargon of our barbarous ancestors,"19 writes Johnson. But after the introduction Christianity, the Angles, Saxons, and Jutes

became by degrees acquainted with the Roman language, and so gained, from time to time, some knowledge and elegance, till in three centuries [after 450] they had formed a language capable of expressing all the sentiments of a civilised people.20
Thus after the language "began to be adapted to civil and religious purposes," it became "artless and simple, unconnected and concise."21 The Norman Invasion brought about the next phase: the terms of philosophy and science entered the language along with an infusion of Norman French. Few in the eighteenth century apart from Thomas Warton had anything to say about the poetry of the eleventh through fourteenth centuries, but there was far more interest in the final stage of refinement, which began in the late fourteenth century: "From the time of Gower and Chaucer," writes Johnson, "the English writers have studied elegance, and advanced their language, by successive improvements."22

By the end of the sixteenth century, the process had reached its peak, and from the writers of that time a pure and copious language may be collected:

From the authours which rose in the time of Elizabeth, a speech might be formed adequate to all the purposes of use and elegance. If the language of theology were extracted from Hooker and the translation of the Bible; the terms of natural knowledge from Bacon; the phrases of policy, war, and navigation from Raleigh; the dialect of poetry and fiction from Spenser and Sidney; and the diction of common life from Shakespeare, few ideas would be lost to mankind, for want of English words, in which they might be expressed.23
Here are the same claims for Renaissance English that were made by the humanists for Augustan Latin, and also, less successfully, for Saxon and Restoration English: it plays the part of a pure native tradition. It retains the vigor of the Gothic ages but anticipates the refinement of Temple, and offers just the kind of classical purity the eighteenth century sought.
A brief excursus. When the Renaissance began to become England's golden age, its language was for the first time allowed to have a character of its own: to be more than a step on the way to something else. It achieved "a stile which never becomes obsolete, a certain mode of phraseology so consonant and congenial to the analogy and principles of its ... language as to remain settled and unaltered."24 It was therefore for the first time worthy of an attention not merely antiquarian. So, as Boswell writes, Johnson "disapproved of Lord Hailes, for having modernised the language of the ever- memorable John Hales of Eton. ... 'An authour's language, Sir, (said he,) is a characteristical part of his composition, and is also characteristical of the age in which he writes.'"25 This is how he justifies, for instance, the preservation of forms like hugger-mugger in his Shakespeare edition: "That the words now replaced are better, I do not undertake to prove; it is sufficient that they are Shakespeare's: if phraseology is to be changed as words grow uncouth by disuse, or gross by vulgarity, the history of every language will be lost; we shall no longer have the words of any authour; and ... we shall in time have very little of his meaning."26 "The history of our language," he says elsewhere, "and the true force of our words, can only be preserved, by keeping the text of authours free from adulteration."27

This concern with a historically specific state of the language was new to the eighteenth century, but not confined to Johnson.28 Hughes, for instance, recognizes that the language of the English Renaissance had a distinct character, one that was worth preserving, in producing the first old-spelling edition of a modern English text. He resolves not only

to preserve the Text entire, but to follow likewise, for the most part, the old Spelling. This may be thought by some too strict and precise; yet there was a Necessity for it, not only to shew the true State of our Language, as Spenser wrote it, but to keep the exact Sense.29
Johnson carefully preserves the original orthography in his "History of the Language," though not in his illustrative quotations. The Six Old Plays published in 1779 are likewise in old spelling. Jonathan Richardson goes so far as to argue that the 1674 edition of Paradise Lost "ought to be the Model of Some Future Edition, and follow'd Letter for Letter and Point for Point."30
So much for the digression; now to conclude. In the larger study of which this is a part, I explore what the age of Elizabeth meant to the age of Johnson, and try to examine it from a number of angles. One thing all these perspectives have in common is the way in which the period we now call the Renaissance was beginning to coalesce into a coherent age -- not only with its own identity, but, more important, with its own cultural uses. Johnson's age, that is to say, used the previous age to understand, authorize, regulate, even create itself, just as the humanists situated themselves in history by referring to their beloved antiquity and despised Middle Ages. The same thing was happening, I have argued here, with the language. As Elizabeth's age began to assume a coherent identity, writers looked backwards to understand and regulate their language: they treated a privileged point in history as a norm and regarded any deviation from it as barbaric corruption. The sixteenth century's insecurity about its Latinity was eased only by appeals to Cicero and his contemporaries; the eighteenth century's insecurities about its language likewise turned to Shakespeare and his. Eighteenth-century linguistic histories entangle these historical judgments with logical and aesthetic judgments so tightly, in fact, that it is difficult to separate them. Languages, it seems, need such privileged moments, for golden ages can cut many of the Gordian knots that appeals to logic or beauty cannot untie. Johnson's own attention to "false refinement and declension," with its implication of a falling-off from a linguistic golden age, reveals one of the most important ways Johnson himself integrates historiography into considerations of the English language, hoping to "mak[e] our ancient volumes the ground-work of stile, admitting among the additions of later times, only such as may supply real deficiencies."


1. Dictionary, sig. C2r.

2. See, for instance, Robert Estienne's Dictionarium, seu Latinae linguae Thesaurus, non singulas modo dictiones contenens, sed integros quoque Latine et loquendi et scribendi formulas, ex Catone, Cicerone, Plinio, Avunculo, Terentio, Varrone, Livio, Plinio Secundo, Virgilio, Caesare, Columella, Plauto, Martiale (1536).

3. Eunuchus, Prologue, 41.

4. Rambler 208; V, 318-19.

5. History of the Royal Society, p. 113. Pellisson-Fontanier too has linguistic purity in mind when he discusses the founding of the Academie Françoise: Richelieu "a eû un si grand succez de cette institution, qu'il a vû la langue Françoise abondamment purifiée" (Histoire de l'Academie Françoise, pp. 4-5).

6. "Defence of the Epilogue"; Prose Works, IV, 9-10.

7. Swift, Prose Works, II, 174.

8. Prose Works, IV, 8-9.

9. Letter to The World, in Johnson: The Critical Heritage, pp. 96-97; Letters, V, 2097.

10. Dictionary, sig. C2r; Idler 63, p. 197.

11. Dictionary, sig C3r.

12. McLaughlin, Literary Imitation in the Italian Renaissance, pp. 262-74. McLaughlin argues that Bembo is in fact the first to "impos[e] the humanist Latin model of Golden Age-Decline-Revival" on vernacular literature (p. 268). The Preface "A Lettori" reads: "abbiamo stimato necessario di ricorrere all' autorità di quegli scrittori, che vissero, quando questo idioma principalmente fiorì, che fù da' tempi di Dante, o uer poco prima, sino ad alcuni anni, dopo la morte del Boccaccio." Writers later than 1400, he complains, "corruppero non piccola parte della puritů del sauellare di quel bon secolo" (sig. a3v).

13. Dictionary, sig. D1r.

14. It was, however, known to Laurence Nowell in the sixteenth century and to Robert Cotton in the seventeenth. Thorkelin transcribed the Beowulf manuscript in 1786 and 1787, and his results appeared as De Danorum rebus gestis secul. III & IV: poema danicum dialecto anglo-saxonica (Copenhagen, 1815).

15. Mary Segar, "Dictionary Making in the Early Eighteenth Century," p. 210.

16. Joseph Spence, Observations, p. 170.

17. New World of English Words, sig. c2v.

18. Works, III, 210-11.

19. Idler 63; p. 197.

20. Dictionary, sig. D1r. Compare Benjamin Martin on Anglo-Saxon: "tho' it was not quite pure, it contain'd but a small admixture of the British and the provincial Latin" (Institutions of Language, p. 13).

21. Idler 63; p. 197.

22. Idler 63; p. 198.

23. Dictionary, sig. C1r.

24. Preface, in Shakespeare, VII, 70.

25. Life, IV, 315.

26. Shakespeare, VIII, 996.

27. Preface to Shakespeare; Shakespeare, VII, 105.

28. Note that antiquarians even in the early seventeenth century had been careful to preserve older orthography in their transcriptions of medieval manuscripts and inscriptions, as in Weever's Ancient Funerall Monuments. But works in English published after roughly 1500 had never received similar treatment.

29. Hughes, The Works of Mr. Edmund Spenser, I, xci. Compare this to the 1719 edition of John Donne, which sometimes resorts to old spelling -- "Goe, and catch a faling starre" -- but just as often regularizes the orthography. The intention here seems to be quaint archaizing, not painstaking transcription of old spelling.

30. Explanatory Notes and Remarks, p. cxxxviii. The effort that evidently went into Hughes's glossary of 777 Spenserian words also reveals a desire to historicize the language, as does Sewell's glossary of 174 "Old Words" in Pope's edition of Shakespeare. When Fairfax's translation of Tasso's Jerusalem Delivered was reprinted in 1749, it came with a six-page glossary of words such as "brast," "kest," and "nathless."