University of Virginia Library

The Text of the Spectator
by
Donald F. Bond

Of the many collections of eighteenth-century periodical essays none has been more frequently reprinted than the Spectator of Addison and Steele. By 1729, the year of Steele's death, at least twelve "editions" had been published by Tonson, and throughout the eighteenth century a steady stream of reprints and so-called new editions emanated from the presses of London, Edinburgh, and Dublin. From 1789 to the present six notable full-dress versions have appeared, with notes and some effort at editorial supervision. Yet, like the majority of eighteenth-century classics, there is no text of the Spectator in existence which enables us to read with even approximate accuracy the essays in a form approaching the intentions of their authors.

The first text with any editorial supervision is the so-called Percy edition of 1789. Thomas Percy began to collect materials for this edition in 1764. On 24 April of that year he wrote to Farmer:

It is clear from this letter and Percy's other correspondence that the chief concern in the new edition was that of providing explanatory notes and not that of a correct text. Percy eventually gave up the work when he became chaplain to the Duke of Northumberland, and his materials were turned over to John Calder, who with John Nichols ultimately brought out the edition in 1789. The value of this edition lies in its biographical and historical notes, not in its text, and the same is true of the two editions which followed at a short interval of time, the first by Robert Bisset in 1793, "with notes and lives of authors," and the other by Alexander Chalmers in 1806.

It was not until 1868, with the appearance of the edition by Henry Morley, that any attempt was made to clear the text of the inevitable corruptions and "improvements" which had crept in with the numerous reprintings of the essays. At that time Morley estimated, on the basis of a sampling of forty numbers, that there were some three thousand textual corruptions in the Spectator. Many of these he deserves credit for having removed, though they have been reintroduced by subsequent editors. Though he is often vague in his account of the texts, Morley seems to have looked at the three original versions, and he sometimes, though not consistently, indicates variant readings in his footnotes. His edition remains to-day the most informative on textual matters.

The only other editions with any pretensions to scholarly exactness are the two which appeared almost simultaneously in 1897 and 1898—those prepared by G. A. Aitken and G. Gregory Smith. Of these Aitken's is modernized in spelling, punctuation, and capitalization, whereas Gregory Smith's professes to reproduce exactly such matters of eighteenth-century "style," even to the insertion of new catchwords at the bottom of the pages. Both announce themselves as based on the first 8vo edition, with occasional footnotes recording variant folio readings, and both ignore the first 12mo edition. Of the two, Aitken's is the less inaccurate, in spite of its modernization of style. The edition of Gregory Smith, the last volumes of which were brought out under pressure of time, reproduces in a strange manner errors to be found only in mid-eighteenth century reprints of the Spectator, and it introduces many other misreadings of its own.[2] This edition was reprinted in "Everyman's Library" in 1907, and in 1945 "reset with minor revisions," only a very small percentage of the errors being set right. Aitken's edition follows consistently, indeed slavishly, the first 8vo text, even when this is patently in error. Neither Aitken nor Gregory Smith appears to have made a thorough collation of texts.

The three earliest and most important texts are the original folio sheets, the first 8vo edition, and the first 12mo edition. The original 555 numbers of the Spectator, printed on both sides of a folio half-sheet and published daily, except Sunday, from 1 March, 1711, to 6 December, 1712, came from

the shops of Samuel Buckley and Jacob Tonson, Junior, which seem to have worked in pretty regular alternation, in order to get out the three thousand or more daily numbers which were called for.[3] A continuation, in which Steele had no share (numbers 556-635), was published by Buckley and Tonson thrice a week from 18 June to 20 December, 1714. The present study is concerned only with the original series. Before the Spectator had gone very far, the popularity of the essays invited a reprint in book form, and a subscription edition was undertaken by Tonson. The first two volumes of this edition in octavo (comprising the first 169 Spectators) was announced in No. 269 (8 January, 1712) as ready for delivery. Ten days later, on 18 January, Volumes I and II were advertised in No. 278 as published in duodecimo, "a neat pocket edition" evidently designed for popular consumption, although the price of the volumes does not appear. Volumes III and IV, again in both 8vo and 12mo (comprising numbers 170-321), were published towards the end of 1712, about a month before the original series ceased daily publication. The 12mo edition is advertised in No. 533 (11 November, 1712) as "this day" published, and the 8vo edition seems to have come out about the same time, although there is no advertisement for this in the original sheets.[4] If the pattern of publication followed that of the first two volumes we may infer that Volumes III and IV appeared in 8vo about the first of November and in 12mo, as we have seen, on the eleventh. Volumes V, VI, and VII, completing the original series, appeared in both 8vo and 12mo on 11 April, 1713; and the continuation of the Spectator (comprising numbers 556-635) was published as Volume VIII, again in both 8vo and 12mo, on 1 September, 1715.[5]

These three earliest editions—the Folio, 8vo, and 12mo versions—offer themselves as candidates for a basic copy-text. In addition, the reprint of Addison's eighteen Milton papers, published by Tonson in 1719, the year of Addison's death, under the title, Notes upon the Twelve Books of Paradise

Lost, collected from the Spectator, contains numerous stylistic revisions and clearly possesses authority. The edition of Addison's Works, prepared by Thomas Tickell and published by Tonson in 1721, is not of great importance. Although Tickell had been designated by Addison as editor and the selection of material seems to have been made by Addison, the text of the Spectators reprinted there is demonstrably inaccurate and can in no sense be taken as the final version intended by the author. The editions, then, which have to be considered as possibly authoritative are the Folio sheets, the first 8vo edition, and the first 12mo edition, together with, for Addison's Milton essays, the 1719 collection of the papers on Paradise Lost.

One has only to place the three texts of any number of the Spectator side by side to see that they differ widely, not only in punctuation, spelling, and capitalization, but in phrasing, in grammatical construction, and in literary style. For how many of these changes—in spelling, for example—were Steele and Addison responsible? How many were due to the style of the printing-house, or to the care—or negligence—of the compositor? Inadvertent errors disfigure all three texts, and many of those in the first 8vo reprint have been carried into modern editions. In No. 407, for example, the word "Pack-thread" is misprinted in 8vo as "Pack-threak," an error retained by Gregory Smith in his edition of 1897-98; in No. 376 the "Scream" of the criers in the streets is altered in 8vo to the meaningless "Stream," a substitution copied by both Gregory Smith and Aitken, and still uncorrected in the 1945 Everyman's Library reprint of Gregory Smith's edition. Some typographical errors, such as "impertiently" in No. 91 of the Folio text, remain in both 8vo and 12mo; others, such as "Inablity" (No. 151, Folio), are followed in 8vo but corrected in 12mo; still others, such as "preceds" (No. 285, Folio), are corrected in 8vo but allowed to stand in 12mo. Obvious errors of this sort can easily be noted of course and corrected, but other examples are less clear, In No. 290, for instance, the Folio reads "in Greatness of Sentiment," which the 8vo alters to "in greatness of Sentiments," a change from singular to plural followed by Morley, Aitken, and Gregory Smith. Again, in No. 310 the Folio reads "in like Manner," the 8vo "in a like Manner": Morley adopts the first reading, Gregory Smith and Aitken the second. Are the 8vo alterations in both cases instances of revision or of inadvertent error? Neither of the readings can be said to commend itself instantly on purely literary grounds. It soon becomes clear, in fact, that no one of the three earliest editions can be followed uncritically, because all contain errors in varying degree.

The relationship of the three texts cannot be represented as the simple one of Folio copied by 8vo and 8vo copied by 12mo, but rather is one of

considerable complexity. Collation of all the variants in the 555 numbers reveals that in certain portions of the text the 8vo and 12mo versions are indeed so identical as to imply complete dependence of 12mo upon 8vo; in other areas, however, the 8vo follows the Folio with little or no change, and the 12mo exhibits considerable variation; and in still other sections the 8vo deviates greatly from the Folio, with the 12mo agreeing with the Folio very closely. Since Tonson and Buckley not only worked alternately in the printing of the original Folio sheets but also, as we shall see, divided between them the work of bringing out separate volumes of both the 8vo and 12mo editions, a study of printing variants in the three texts of the Spectator offers an unusual opportunity of discovering to what extent printing practice determined changes in spelling, punctuation and the like, and how far these seem to have been due to the author. It will be of interest to apply here the theory, so persuasively stated by Greg, that the first printing of a text—which is alone set up from an author's manuscript—represents most faithfully the writer's intentions and that the farther away we get from the first printing the greater will be the opportunity for inadvertent changes and compositors' normalizations to occur.[6] Greg's conclusion is that for the "accidentals" of a text—spelling, punctuation, and capitalization—the earliest edition set from an authoritative manuscript should be chosen as copy-text, with the insertion of such "substantive" or other alterations from the revised editions which in the judgment of the editor are authoritative. This procedure, which is coming to be pretty generally recognized in studies of sixteenth- and seventeenth-century texts, seems particularly relevant to an inquiry into the text of such a work as the Spectator, printed first in Folio sheets and later revised for publication in book form.

Through the greater part of Volume I (Nos. 1-80) the 8vo and 12mo texts of the Spectator are quite similar. In numbers 12, 13, and 19 the two texts are identical so far as accidentals are concerned, and this is generally true through No. 73, with the 12mo copying typographical errors of 8vo in numbers 15 and 28. There seems little question but that for most of the volume the copy-text used by 12mo is the printed 8vo sheets. For the last seven numbers, however, there is considerable divergence, with 12mo following in about equal proportions Folio and 8vo readings, and in No. 80 copying an error in Folio. For Nos. 74-80, therefore, we may conclude that the 12mo text is less insignificant than it appears to be in the earlier and greater part of the volume.

In Volume II (Nos. 81-169) the pattern is much the same as in the

first part of Volume I, with the 12mo text closely following the 8vo. A substantive revision, however, appears as early as No. 91, and others follow in Nos. 101, 104, 110, 119, 125, 131 (as many as seven in this number), and so on, in increasing numbers in the latter half of the volume. The 12mo text shows little similarity to the Folio throughout Volume II, and there is no instance of its copying errors in the Folio text. It does reproduce errors, however, from the 8vo, so that one must conclude that its copy was the 8vo sheets. The increasing number of revisions, on the other hand, suggests that the authors may have taken the opportunity to make further corrections after the 8vo sheets had been printed.

The 8vo and 12mo texts diverge radically in Volume III (Nos. 170-251), with the 8vo retaining many of the Folio characteristics and the 12mo showing for the most part independent readings, so far as accidentals are concerned. In No. 221 the 8vo text follows the Folio in every detail, whereas there are 38 changes in 12mo. In only four (Nos. 170, 178, 184, and 251) of the eighty-two papers in this volume is the agreement between 8vo and 12mo greater than that between Folio and 8vo. The 8vo copies errors of the Folio in Nos. 202, 212, 223, 228, 234, 241, and 249; in only one instance (No. 242) is an error in Folio copied by 12mo. The 12mo text shows a similar independence as regards substantive readings and in as many as twenty-seven essays contains authorial revisions not to be found in 8vo. Significantly it is only in this volume that we find an instance of an erratum marked in the Folio and corrected in only one of the reprints. In No. 235 the substitution of "Fence" for "Force" is noted only in the 12mo. It can hardly be argued, on the other hand, that the 8vo is following an uncorrected copy of Folio, for a considerable share of substantive revisions are to be found in both 8vo and 12mo. These often occur in slightly different form, so that one infers that both reprints are using independently a corrected Folio sheet. One example may be given (from No. 249):

In Volume IV (Nos. 252-321) the normal pattern of concurrence between 8vo and 12mo is resumed, but only as far as No. 270. Beginning with No. 271 the 12mo text suddenly begins to show a preponderance of identical readings, both accidentals and substantives, with Folio. More precisely, the shift occurs not at the end of No. 270, but about midway in the fourth paragraph of No. 271, at the conclusion of gathering H of the 8vo edition. From the beginning of gathering I of the 8vo to the end of the volume the 12mo text, especially as far as accidentals are concerned, is much closer to the Folio than to the 8vo. In No. 305, a fair example, the 12mo agrees with the 8vo in five cases, with the Folio in sixty-one. It seems likely that the printing of 12mo was proceeding at a faster pace than that of 8vo and that the supply of 8vo printed sheets, which the 12mo compositor had been using for copy, were no longer available, so that from No. 271 the Folio sheets became the copy for 12mo. As we should expect, 12mo copies errors from 8vo in the early part of the volume (in Nos. 265, 266, and 270), and after No. 270 copies errors from Folio (in Nos. 272, 278, and 285). Near the end of the volume, however (in No. 319), it repeats an error from 8vo, in the spelling "disigenuous," which, unless due to coincidence, suggests the following of 8vo.[7] The dependence of 12mo on Folio need not in itself, however, imply an inferior state of text for 12mo; in several of the papers in the latter portion of the volume (notably in Nos. 272, 275, 279, 285, 298, 309, and 317) it preserves correct substantive readings which had been lost in the 8vo. On the other hand, the correction of errata in two numbers falling in the latter half of Volume IV seems to point to further revision induced by "second thoughts" in the preparation of 8vo. The erratum for No. 283 has "for Catlaine's read Cataline's." The 12mo text duly prints "Cataline's," but the 8vo has the correct spelling "Catiline's." In No. 293 we have the case of two errata originally marked in the Folio sheet but apparently superseded by further revision before the number came to be reprinted.

FOLIO

8vo

12mo

FOLIO

8vo

12mo

If we were confronted with only the second of these cases we might infer that 12mo was using an uncorrected Folio sheet and had only the original errata notice to follow. In the first example, however, both 12mo and 8vo adopt the same revision, and consequently it would appear either that the 12mo compositor was careless or that he had access to only a partially revised sheet. The entire 12mo text of this number is very close to the Folio in accidentals, and, in contrast to 8vo, fails to correct the spelling "Richlieu" earlier in the paper as well as the incorrect marking of the Greek motto at the head of the essay. Clearly, the 12mo readings in the latter half of Volume IV, standing so close as they do, both in accidentals and substantives, to the Folio, afford a valuable check on the 8vo text, but they offer no easy solution and do not allow an editor to shirk the responsibility of deciding each reading on its own merits.

For the first part of Volume V (Nos. 322-394) the 12mo text continues to show considerable dependence on the Folio sheets, and copies errors from the Folio in No. 326; but after No. 351 the similarity between the two decreases, with the exception of one number (No. 386), and from No.

352 to the end of the volume the 8vo and 12mo texts are in general agreement. Errors in 8vo are copied by 12mo in Nos. 354, 359, 361, and 369. Both 8vo and 12mo of this volume were published simultaneously, and it seems likely that at the beginning each volume was set up from corrected Folio sheets, but that after some thirty numbers had been printed the supply of 8vo sheets began to be available as copy for 12mo. There is a notable absence of independent substantive readings in 12mo. Volume V was produced, both in 8vo and 12mo, by Buckley, and the proportion of variants is high, particularly in the papers originally printed in Folio by Tonson. It is perhaps worth noting that in the Milton papers, most of which occur before No. 352, the instances in which 12mo follows Folio rather than 8vo are to be found mainly in the quoted passages. Was the 8vo compositor using a separately marked copy of Paradise Lost and the 12mo compositor simply transcribing the Folio sheets?

Volume VI (Nos. 395-473) is notable for the low number of accidental variants, and there is a greater uniformity among the three texts than in any of the other volumes. In No. 413 there are only seven variants, a proportion of only 4% to the length of the paper. This is an extreme case, but in only eleven numbers of this volume does the proportion rise above 20%, in remarkable contrast to Volume V, in which all but two of the papers show a figure higher than 20%. Generally the 8vo and 12mo texts agree; only in three numbers (Nos. 411, 439, and 468) is there a greater correspondence between 12mo and Folio. Judging from the reproduction of errors, the 12mo text seems to follow now the 8vo (in Nos. 400, 403, 422, 425, 427, 450, 465, and 470), now the Folio (in Nos. 407, 413, 436, and 458) as copy. Independent substantive readings are rare in the 12mo, but as usual it frequently preserves correct readings from the Folio.

The proportion of variants in Volume VII (Nos. 474-555) is higher, particularly in the first half of the volume. It is apparent from typographical evidence that this volume, both in 8vo and 12mo, was divided between the two printing houses, Buckley producing the first half of the volume (Nos. 474-517)—through gathering Q in the 8vo and gathering H in the 12mo—and Tonson doing the remainder.[8] In the portion printed by Buckley there is considerable variation between 8vo and 12mo; in the remainder of the volume the 12mo text follows 8vo fairly consistently. The most striking feature is the fact that in the first half of Volume VII the proportion of variants is much higher (an average of 36% per number) in the papers which Tonson had originally printed in Folio than in the papers originally produced by Buckley himself (an average of 21%).

Conversely, in the latter half of the volume the average of variants for the papers originally printed by Tonson is only 12%, whereas the average for those originally brought out by Buckley is 17%. These figures, which can be supplemented by comparisons from other volumes, afford good evidence that changes in accidentals are likely to be due not to the authors of the essays but to the practice of printers.

This survey of the textual variants in the seven volumes enables us to see more clearly the relationship of the three texts. It is apparent that for a large portion of the essays the 12mo text is following the reprinted 8vo sheets. This is true in the main for Volumes I, V, and VI, where 12mo generally reproduces the text of 8vo quite closely. These portions of the Spectator are less likely to offer, in the 12mo volumes, independent and revised readings. The remaining four volumes, particularly Volume III, display radical divergences in the 12mo text, and are entitled to more scrutiny than they have hitherto been accorded. The differences may be summarized by showing for each volume the number of essays in which independent 12mo readings are more numerous than agreement of 12mo with either 8vo or Folio.

     

Variants	I	II	III	IV	V	VI	VII
Accidentals	3	27	79	8	0	1	11
Substantives	0	8	34	13	0	5	11

Since the 12mo text has been almost completely ignored in the past, it is obvious that for certain volumes it may offer the opportunity for new and good readings.

The point raised by the study of variants in Volume VII offers the possibility of exploring the real cause for accidental variants. From the evidence supplied there, it would seem likely that we should find fewer variants of this sort in a number of the Spectator produced by the same printing-house in Folio, 8vo, and 12mo, than in a number brought out in Folio by Printer A and reproduced in 8vo and 12mo by Printer B. Three other volumes of the Spectator enable us to test this hypothesis—I and VI, printed both in 8vo and 12mo by Tonson, and V, printed in both 8vo and 12mo by Buckley. A count of the accidental variants—excluding the substantive changes, for which the author, not the compositor, would be responsible—shows conclusively that an essay originally produced by one printer and reprinted by another contains a greater number of variants. In Volume I (published in 8vo and 12mo by Tonson) the number of accidental variants in the 40 essays which Tonson had originally printed in Folio is 1,123; for the same number of essays first printed by Buckley the

total is 1,287, an increase of 12½%. In Volume V (by Buckley) the variants in the 36 essays originally printed by Buckley total 1,551; the variants in the same number of essays first printed by Tonson total 2,410, an increase of over 35%. The Tonson papers in this volume are considerably longer, with a total "lineage" almost 25% greater than that of the Buckley papers. Even with this modification, however, the increase is sufficiently impressive. In Volume VI (by Tonson) the 39 papers originally done by Tonson show a total of 501 variants; in the same number of original Buckley papers the total reaches 996, an increase of almost 50%. The total length of the Buckley papers in this volume is only 9% greater than that of the Tonson papers. The variants when classified according to the authors of the essays do not show the same consistent proportions. It is the printer who is responsible. Variants in spelling, punctuation, and capitalization exist not because the author of the essay demanded these changes but because the preferences of the compositor or the practice of the printing-house dictated a different spelling, a heavier or lighter punctuation, or a more consistent system of capitalization.

The evidence from spelling is conclusive. In the dedication of Volume II to Lord Halifax the first edition of the 12mo (Sig.A2v) reads: "While I busie my self as a Stranger upon Earth, and can pretend to no other than being a Looker-on, You are conspicuous in the Busie and Polite World. . . ." In the second edition the spellings "busy" and "Busy" are substituted—not because Steele rewrote the dedication, but because the first edition was produced by Tonson, the second by Buckley. A more arresting example occurs in two of the Folio Spectators (Nos. 170 and 171), both written by the same author (Addison) on the theme of jealousy but printed by Buckley and Tonson respectively. In the first essay "jealousy" is spelled without exception (twelve times) with the "y" ending, and in the following paper as regularly and uniformly (six times) with the "ie" termination. It is unthinkable that Addison wilfully altered the spelling of the key word in the two essays, spelling it "jealousy" throughout the one and "jealousie" throughout the other. When Tonson reprinted these essays in the 8vo edition, the spelling throughout both numbers is "jealousie," whereas the 12mo edition of Buckley with almost equal uniformity (there are two exceptions) spells the word "jealousy." The responsibility for the changes is unmistakable. The letters of Addison and Steele exhibit the greatest irregularity in spelling,[9] and one conjectures

that the Spectator manuscripts confronting the compositor showed a similar lack of consistency. Since the Spectator was produced by two printing-houses, both in the original printing and in the 8vo and 12mo volumes, it offers a particularly good source for evidence of spelling differences. The two spellings of "busy" have been cited. In the Folio sheets printed by Tonson the word is generally spelled "busie," but it is never spelled thus in the Folio numbers printed by Buckley. In the Folio, again, the word "battle" appears uniformly thus in numbers by Buckley, and without exception as "battel" in numbers by Tonson. The table below brings together the most clearly marked differences in spelling, in the Folio sheets, the 8vo, and the 12mo edition. The first column lists all the occurrences of each spelling in the 555 Folio numbers, first in the papers printed by Tonson and secondly in those printed by Buckley. In the seven columns following will be found the corresponding spellings of each word in the seven volumes of the reprint, first in 8vo and then in 12mo. Volume VII is divided, to show (a) the portion (Nos. 474-517) printed by Buckley and (b) the portion (Nos. 518-555) printed by Tonson. At the foot, for each of the 8vo and 12mo volumes, will be found the total of characteristic "Buckley" and "Tonson" spellings.

                                         

	FOLIO	I		II		III		IV		V		VI		VII
	T	B	8°	12°	8°	12°	8°	12°	8°	12°	8°	12°	8°	12°	8°	12°
															(a)	(b)	(a)	(b)
Battle	0	34	0	2	5	2	0	3	6	3	6	5	2	1	4	0	4	0
Battel	53	0	18	16	3	6	4	1	7	10	16	17	14	15	0	2	0	2
busy	3	25	0	0	8	3	1	6	6	0	2	4	0	0	2	0	2	0
busie	18	0	6	6	2	7	6	1	1	7	2	0	8	8	1	3	1	3
Centre	2	6	0	0	1	1	0	1	2	2	0	0	4	4	1	1	1	1
Center	10	1	0	0	0	0	1	0	0	0	6	6	0	0	3	0	3	0
easy	12	124	0	0	13	3	4	26	26	0	18	23	0	0	10	0	12	0
easie	66	3	19	19	6	16	29	7	9	35	5	0	40	40	4	23	2	23
extreme	8	61	3	3	6	5	10	12	10	9	2	7	5	6	5	2	4	2
extream	45	13	12	12	14	15	11	9	8	9	17	12	14	13	2	6	3	6
humane	9	99	0	0	11	4	15	46	29	11	19	20	2	3	19	2	23	4
human	110	10	15	15	15	22	43	12	2	20	11	10	40	39	6	31	2	29
Money	22	87	0	0	16	5	16	30	33	25	10	8	0	0	10	1	11	0
Mony	63	0	19	19	5	15	16	2	2	10	0	2	30	30	1	12	0	13
perswade	12	47	4	4	4	1	12	14	12	12	6	6	7	8	11	3	11	3
persuade	12	0	2	2	0	3	4	2	3	3	0	0	3	2	0	2	0	2
-our	42	104	0	0	21	13	9	14	20	19	15	15	19	20	10	3	6	1
-or	115	58	44	44	18	26	34	29	33	34	24	24	28	27	24	29	29	20

             

-y (verbs)	61	102	7	7	15	8	6	12	19	7	22	31	7	7	13	18	18	20
-ie	135	79	51	51	23	30	28	22	32	44	30	21	61	61	20	26	17	22
-y (nouns)	9	49	0	0	5	3	1	29	5	0	6	10	0	0	5	1	6	1
-ie	38	5	7	7	3	5	35	7	1	6	4	0	21	21	1	4	4	0
			--	--	--	--	--	--	--	--	--	--	--	--	--	--	--	--
"Buckley" spellings			14	16	105	48	74	193	168	88	106	129	46	49	90	31	98	32
"Tonson" spellings			193	191	89	145	211	92	98	178	115	92	259	256	62	138	61	120

The distribution of spellings is curious, and cannot be due to coincidence. The word "money," for example, which is never spelled as "mony" in the Folio sheets printed by Buckley, has this spelling in both formats of Volume I and Volume VI and is never spelled thus in Volume V, 8vo. Some of the spellings—particularly the "ie" and "our" endings—may of course in some cases be due to the need for justifying the line of type, and due allowance must be made for accident or error on the part of the compositor. Most of the differences, however, occur in sufficient numbers and with enough consistency to rule out the element of chance. A comparison of the spellings in the Folio sheets with those in the following columns points very strongly to the printers responsible for the 8vo and 12mo volumes: the 8vo volumes I, III, and VI, and the 12mo volumes I, II, IV, and VI by Tonson; the 8vo volumes II and IV, and the 12mo volumes III and V by Buckley; with Volume VII, both in 8vo and 12mo, clearly divided between the two. Only for Volume V, 8vo, is the evidence inconclusive, and on the basis of typography it may be assigned to Buckley. For all the other volumes the typographical evidence confirms the data from spelling differences, and the several volumes may provisionally be assigned as follows:

                 

OCTAVO	DUODECIMO
I. Tonson	Tonson
II. Buckley	Tonson
III. Tonson	Buckley
IV. Buckley	Tonson
V. Buckley	Buckley
VI. Tonson	Tonson
VII. (a) Buckley	Buckley
(b) Tonson	Tonson

Our concern here, however, is not so much to prove the identity of the printers responsible as to note the remarkable consistency of spelling differences and the very strong support which this gives to the hypothesis that these differences are attributable not to the authors of the essays but to the compositor or the printing-house. And what is true of spelling is, I feel sure, also true of other accidentals such as capitalization, punctuation, italicizing, abbreviating, and so on. The Spectator was published at a time when these matters were in process of becoming standardized, but there was still considerable latitude. McKerrow has pointed out how different, in the matter of capitalization, is Rowe's text of Shakespeare, printed in 1709, from that of Pope, published just sixteen years later: If sufficient manuscript material by Addison and the other authors of the essays existed it might be possible to attempt to determine the general practice of these authors and proceed to standardize spelling and other accidentals accordingly. Such material is not available, and any attempt to regularize the Spectator essays to fit such a program is open to the two-fold objection that it would necessarily be incomplete and that it would impose a far greater uniformity in such matters than Steele or Addison or any of the other contributors ever cared about or practised. A choice among copy-texts, however, must be made, even though such a text will certainly exhibit a number of non-authorial characteristics. The evidence assembled here gives very strong support to Greg's view that the first printing—in the case of the Spectator the Folio sheets—will be freest from contamination and the closest in the presentation of accidentals which we can hope to reach to the original intentions of the authors. The later corrections and changes which were made by the authors—substantive revisions—will then be incorporated into this basic text.

In the Spectator these revisions may be said to begin with the lists of errata in the Folio sheets. As early as the sixth paper an errata list is provided for No. 5, and these occur at intervals throughout the run of the Spectator—a total of 57 errata lists in the 555 numbers, distributed among 38 numbers by Addison, 17 by Steele, one (No. 283) by Budgell, and one (No. 237) of uncertain authorship. Some of these errata have to do with improvements in style, some are corrections of fact, and some are simply rectifications of spelling. Many factual errors and misspellings remain unnoticed, however, so that it is clear that the insertion of errata lists was not done in any very systematic manner. The greater part of them have to do with the Milton papers, and in the later numbers of the Spectator the occurrence of errata lists diminishes perceptibly.

The errata are, with some exceptions, taken into account with the reprinting of the essays in 8vo and 12mo. In addition, many stylistic changes are made—a total of over 1,500. They are unmistakably author revisions. Addison, for instance, makes a number of stylistic alterations to avoid inadvertent repetition of a word, and with fairly regular consistency he drops the relative pronoun "that" and replaces it by "which" or "who." Addison, in fact, pays much more attention to such matters than does Steele and is responsible for more than two-thirds of the total number of revisions. Of the 555 papers only 51, less than one-tenth, are without any stylistic changes; of these 33 are papers by Steele, 14 by Addison, two by Budgell, and two by uncertain authors. Moreover, the number of such revisions in single papers is much higher in Addison's than in Steele's; Addison's papers will contain as many as twelve or fifteen stylistic revisions within an essay. Steele's revisions run to a lower figure: there are a few papers with as many as seven, but the number is generally lower. Only 41 of Steele's papers contain more than three such revisions. Steele's inattention may be illustrated by the manner in which he allows reproduction of his own earlier writings to stand in the Spectator. On two occasions he quotes long extracts from The Christian Hero, in texts which are palpably careless and inexact. The phrase "the Machiavilian Scheme" appears in No. 516, Folio, as "the Machiavilian Scene," an error which is allowed to stand in the 8vo and 12mo texts. Again, the phrase "Their proud and disdainful Hearts, which were putrified with the Love and Pride of this World" appears in No. 356 (all three texts) with the reading "petrifyed." One might argue that this change was Steele's own revision but for the fact that he continues to use "putrified" in editions of The Christian Hero after 1712.

When Steele and Addison make stylistic revisions for the 8vo and 12mo editions, and when we can be sure that the changes are in fact revisions, we are entitled to consider them as representing what the authors of the Spectator intended us to read. Too often in the past, however, editors have been ready to take all such readings in the 8vo text as authoritative revisions. Actually many are printers' errors. For example, in No. 267, the first of the Paradise Lost papers, an essay which Addison carefully revised, the 8vo and 12mo texts read:

If we turn to the Folio text we see that what Addison wrote was "that just and regular Progress," and this is confirmed by the reading in the 1719 volume of the Paradise Lost essays. The Tickell edition of the Works repeats the erroneous "Process," as do most modern editions, including those of Aitken and Gregory Smith. Morley alone looked at the Folio and has the correct reading. A few additional examples may be cited in which the Folio text is to be preferred.

               

FOLIO	8vo and 12mo
No. 94: he distinguishes every Moment of it with some useful or amusing Thought	with useful or amusing Thought
No. 164: that he might encourage her in the pious Resolution she had taken, and give her suitable Exhortations for her Behaviour in it	the pious Resolutions
No. 295: there should not have been a Woman in the County better dressed	in the Country
No. 342: it was in the Woman he had chosen that a Man of Sense could shew Pride or Vanity with any Excuse	with an Excuse
No. 361: It has often supplied the place of the ancient Chorus, in the Works of Mr. ***	in the Words of
No. 455: I was naturally led into a Reflection upon the Advantages of Education, or moral Culture	modern Culture
No. 478: Old Beaus are to be preferr'd in the first Place	to be presented

Another interesting—and much larger—body of examples occurs in which the correct reading of the Folio text has been lost in the 8vo reprint but retained in the 12mo version. This edition, inferior in appearance to the handsome subscription 8vo, has been almost entirely ignored in the past. The fact is, however, that while it introduces errors of its own, it frequently provides, as we have seen, evidence of being independently set up from the corrected Folio sheets, and it avoids many of the errors of the 8vo edition. In No. 290, by Steele, the 8vo text reads:

It would not be difficult to conjecture that for "Imagination" we should read "Examination," and this is, in fact, the reading of both the Folio and 12mo texts. Both Aitken and Gregory Smith retain the 8vo reading. In No. 533, again by Steele, which contains a letter protesting against immoral company in stage-coaches, the 8vo text reads: Turning to the Folio text, we find that the reading is not "silly" but "filthy stupid Questions." The 8vo compositor misread the word, but the 12mo compositor got it right and printed it as it stood in the Folio. All succeeding editors have followed the erroneous 8vo text. I cite a few more examples in which the 8vo text is clearly wrong.

                 

FOLIO and 12mo	8vo
No. 113: I made new Liveries, new paired my Coach-Horses, sent them all to Town to be bitted, and taught to throw their Legs well, and move all together	and move altogether
No. 210: If he considers his Being as circumscribed by the uncertain Term of a few Years, his Designs will be contracted into the same narrow Span he imagines is to bound his Existence.	to bound to his Existence.
No. 224: as it inspires rational Ambitions, correct Love, and elegant Desires.	rational Ambition, corrects Love, and elegant Desire.
No. 250: since the several Treatises of Thumbs, Ears and Noses have obliged the World, this of Eyes is at your Service.	the several Treaties
No. 264: The Hogsheads of Neat Port came safe, and have gotten thee good Reputation in these Parts	the good Reputation
No. 298: They would face me down, that all Women of good Sense ever were, and ever will be, Latitudinarians in Wedlock and always did, and will, give and take what they profanely term conjugal Liberty of Conscience.	profusely term
No. 448: the Expectation which is raised by impertinent Promisers	Promises
No. 456: these Men are to be valued only for their Mortality, and as we hope better Things from their Heirs	Morality

 

No. 521 (on lies and distortions of fact): These and many other Hints I could suggest to you for the Elucidation of all Fictions

Factions

A study of these examples suggests that the Folio text not only in accidentals but in substantive readings possesses more authority than has generally been recognized, and that, although it was revised when the Spectator was reprinted in book form, the compositors of the 8vo and 12mo versions—particularly of the 8vo—altered the text in many cases for the worse. Interestingly enough, in view of the fact that earlier editorial collation seems to have centered largely on the early papers, the errors increase in number in the later volumes: in the first two volumes I find only 17 instances of erroneous substitution in the 8vo, but there are 57 in Volumes III and IV, and 107 in Volumes V, VI, and VII. With the publication of the Spectator in book form these errors became firmly incorporated in the text, and no one, with the partial exception of Morley, seems to have looked at the Folio readings from this point of view or attempted a serious collation of the three texts.

If the Folio text thus merits attention, an examination of the 12mo edition shows that it also needs to be taken into account as a source of substantive readings. While it introduces many obvious errors, it provides some examples of independent stylistic revisions, with readings, consequently, which are not to be found in the Folio or 8vo edition. In Volumes I and II there is not much difference textually between the 8vo and 12mo editions, although in Volume II the 12mo begins to diverge noticeably. In Volume III there are radical differences (a preponderance of differences in 79 out of the 82 numbers in the volume). In Volume IV the differences are much less striking, and in Volumes V-VII they become almost negligible. To put it another way, in the first two volumes there are eight readings which appear to be authorial revisions distinct from the Folio and 8vo versions; in Volumes III and IV there are 47; whereas in Volumes V-VII there are no more than 16. Three examples may be cited. In No. 131 the Folio text, followed by the 8vo, reads: "Some look upon me as very proud, and some as very melancholy." The 12mo version looks like an author's revision: "Some look upon me as very proud, some as very modest, and some as very melancholy." In No. 195, in a paragraph on the long-lived Lewis Cornaro, the Folio text reads:

The 8vo text follows this reading, and Gregory Smith has the following note upon the passage: "The English version incorrectly referred to by Addison is Sure and certain Methods of attaining a long and healthful Life. . ." The 12mo edition, however, contains the correct reading: The third example to be offered involves a drastic change in meaning. In No. 279, one of the Paradise Lost essays, in a passage in which Addison is discussing the sublimity of Milton's thoughts, the Folio text reads: Addison revised this sentence, striking out, I assume, in the Folio sheet the last three words and inserting in the margin, "second and sixth Books." The 8vo compositor noted the correction but apparently misread the word "sixth," so that in the 8vo edition the sentence ends with the words "in his first, second, and tenth Books." Only the 12mo text has the correct reading, "in his first, second and sixth Books." The 12mo reading not only accords better with the facts of Milton's poem but is confirmed by the 1719 reprint of the Paradise Lost papers.

Such are some of the substantive differences in the three early texts. They emphasize unmistakably the dangers inherent in too implicit a reliance upon the 8vo text, the edition which has hitherto served almost exclusively as the source for modern reprints of the Spectator. Studied in connection with the accidental variants they enable us to assess with greater confidence the relative authority of the various portions of the text and to see at first hand some of the processes of change, degeneration, and recovery which operate in the reprintings of a long prose work. The study of variant spellings and conflicting readings not only allows us to restore much that has been lost but also brings some evidence to bear on the responsibility which the printing-house may claim in the texts which we now read. Judged from this point of view the study of variants may escape the censure which the writer of Spectator 470 bestowed upon the textual critics of his own day:

His own prose, which has enjoyed the praise of Dr. Johnson and survived the censure of Bishop Hurd, deserves to be read in as accurate a version as modern scholarship can attain. Such a text, if not imparting "a new Elegance," will at least in many cases allow "a different Sense,"—a sense, one hopes, which its author originally intended and would now approve.

Notes