1. The History of the History
The 1895 paper [Mottram1895] contains a wealth of detail, largely uncontroversial, yet one sentence stands out, and it tenaciously dogs New Zealand philatelic history even to this day. On p158, the author hypothesizes that “It is probable that Mr. Edward Henry Corbould, R.I., made a water-colour drawing of the design [an engraving of the portrait of Victoria] for the first New Zealand stamps for Messrs. Perkins, Bacon & Co.”
We shall see that this supposition is almost surely not the case and, if the Corbould water-colour had any influence, then it was the most minor in nature and really does not justify mention.
[Jolliffe13, p8] does not bite, and merely records “The design … consists of a three-quarter face bust portrait of Queen Victoria … The portrait is from a painting in 1838 of the Queen in her Robes of State by Mr. A. E. Chalon, R. A., a copy of which appears in the frontispiece.”
Nonetheless, Mottram’s supposition became received wisdom in 1916 when Melville wrote that Chalon’s portrait “was the source of the New Zealand stamp portrait of 1855, … Mr Edward Henry Corbould, R.I., is understood to have made a drawing in water-colours from [an engraving of the portrait of Victoria] … from which copy … [the engraver] worked.” [Melville16, p64]
Like so many readers, including this author, R.J.G. Collins savoured [Melville16], and wrote in [Collins22, p89] “One of the finest philatelic handbooks available to the collector is ‘Postage Stamps in the Making,’ by Mr. F. J. Melville”. It comes as no surprise then that Collins defers to Melville’s history when he writes “Popularly known as “full-faces,” the design was adapted from a painting by A. E. Chalon, R.A., in 1838, of Queen Victoria in her state robes. A water-colour reproduction was made by E. H. Corbould, from which the engraver … prepared the ‘mother die’” [Collins22, p8].
On the other hand, a connection between the Corbould water-colour and the New Zealand stamps was not made by Edward D. Bacon when he announced in 1933 that, among other works, he had acquired the Corbould portrait for King George V. The paper included a small black and white reproduction of the miniature [Bacon33, A, Plate I, item A, between p76 and p77]. Instead Edward D. Bacon indicated that the miniature was painted by Corbould for the first stamps of Van Diemen’s Land/Tasmania and was later used for Bahamas, Grenada, Natal and Queensland [Bacon33, p77]. Edward D. Bacon supported this claim via an excerpt from Corbould’s 10 May 1854 correspondence to J.B. Bacon of Perkins Bacon & Co., wherein Corbould writes “I purpose [propose] taking up that which you desire, viz., the Queen’s head for Van Diemen’s Land postage stamp …” [Bacon33, p76]
Other details, and particularly dates, from Corbould’s correspondence to J.B. Bacon reported by [Bacon33, p76-77] are worth repeating since we will use them to build a rebuttal: “In a letter to J.B. Bacon, dated May 10th1854 … Mr Corbould wrote ‘ … in order that I make no blunder respecting size I had better send the little sketch for you to determine the exact proportion, and then if you will let me have it again soon I will make the drawing’ … Five days later on 15 May Corbould wrote again to J.B. Bacon that ‘… I have worked away … to paint it as it ought to be. … If you find any fault of any kind to find with the head, pray send it back and I will try to attend to it.’ … On May 17th Mr [J.B.] Bacon replied ‘I received your note of the 15th inst. yesterday enclosing the drawing of Her Majesty and consider it a perfect gem. I have much pleasure in forwarding a cheque as you suggested …’”
In this author’s view, particularly given the definite article (“the little sketch”), it is likely that Corbould sent a miniature sketch of Victoria on 10 May 1854 so that J.B. Bacon could approve its size, then certainly Corbould sent the miniature water-colour on 15 May 1854, which triggered payment by J.B. Bacon.
[Melville16]’s version of events continued to hold sway with R.J.G. Collins. In the justifiably esteemed [TPSoNZ38v1, p32], Collins writes that “At the request of Messrs. Perkins, Bacon & Co., Edward Henry Corbould, R.I., made a water-colour sketch of the head and bust for the guidance of the engraver … who was entrusted with the cutting of the portrait.”
The rest, as they say, is history, since any New Zealand philatelist seeking authoritative knowledge on the origins of the Chalon heads would naturally refer to a volume of The Postage Stamps of New Zealand, and would naturally consult with the first volume for information about New Zealand’s first stamp issue. Any revision to the consensus history in later volumes would inevitably receive less attention.
|
For instance, eighty years later, almost all of this author's favorite philatelic references would benefit from a modernization of their language in regards to this one point: NZ Post, Virtual New Zealand Stamps, stampsnz.com, previously Wikipedia, and even the description of the 1988 Centenary Philatelic Society in the Campbell Paterson catalogue. Some have addressed the topic already, such as Wikipedia entry (courtesy of this author).
The first volume of the landmark series The History of New Zealand Postage Stamps was edited by R.J.G. Collins and H.T.M. Fathers, then for the second volume C.W. Watts replaced H.T.M. Fathers, and the duo of Collins and Watts continued until the fourth volume. From the Preface of the first volume, R.J.G Collins is listed as contributing the chapter discussed here.
The Corbould painting is described as a water-colour sketch in [TPSoNZ38v1, p32] and [TPSoNZ64v4, p3]. Although the label water-colour is entirely apt, the reference to sketch seems to hint at the 10 May 1854 sketch used for sizing rather than the miniature water-colour drawing sent on May 15. Accordingly, to avoid confusion, this author reserves the term sketch purely for Corbould’s 10 May work.
|
One potential cause for the content in [TPSoNZ38v1] may be a confusion as to the nature of the Corbould miniature. Specifically, the frontispiece of appears to be a half-length portion of a Cousins engraving of the Chalon portrait yet is markedly mis-captioned “Queen Victoria in Her State Robes. From the water-colour sketch by E.H. Corbould, R.I., based on the full-length portrait by Alfred Edward Chalon, R.A. The sketch was followed by the engraver … in cutting the portrait for the first stamps of New Zealand.” Such a misapprehension might come from a mistaken claim or some other source, yet must not have been corrected by a more authoritative reference, such as [Bacon33]. One sentence in the Preface by R. Heaton Rhodes, President of the Philatelic Society of New Zealand, seems to support the latter supposition. As a corollary of making a point about why [TPSoNZ38v1] does not include a list of references, Rhodes also indicates that the authors only had access to a constrained set of literature, via “Unfortunately the philatelic libraries in New Zealand are not sufficiently comprehensive or representative to ensure the completeness of a list of references and the inclusion of some and the omission of others might have given rise to an accusation that distinctions had been made”.
 |
| The caption in the image above mischaracterises Cousin's detailed engraving as the much less detailed Corbould miniature watercolour. Copyright believed expired. |
[TPSoNZ38v1, p33] helpfully supplies more information that we will use shortly in a rebuttal. It describes an essay in the Perkins, Bacon & Co. material where “the portrait was reduced in extent and the network background was cut outside the circle containing the portrait. The inscription NEW ZEALAND and POSTAGE were outlined with the background’s spandrels still showing through, but the value appeared in clear letters. A 1d die in this type was completed on 3 March 1854 and a 2d die on 4 April”.
In 1952, the Corbould miniature water-colour was printed as a colour copy in [Wilson52], albeit grainy at high magnification. Falling quickly on the heels of this sumptuous tome was the publication of the Perkins, Bacon & Co. records in 1953, wherein both the uncleared 1d essay and the Corbould miniature water-colour were reproduced in the greatest of detail [PBR53, plates XI and VIII respectively]. Moreover the author Percy de Worms states “There is no record of Corbould having been employed, nor was any charge made to the colony for the drawing.” [PBR53, Vol. I, p158].
E.H. Corbould is mentioned several times in [PBR53]:
· p.xxvi: “Little appears in the correspondence on the subject of the preliminary drawing, which was largely entrusted first to Henry Corbould and later to his son Edward Henry Corbould, who also did a considerable amount of designing of banknotes for the firm. … At the end of 1874 we find him urging to adopt … drawings made three or four times the size that the engraving is to be … then reduced [to a] photograph. This photo being the exact size required …”
· p53-55: E.H. Corbould might have been involved – or more likely was not involved – in copying drawings of a stamp proposal for New South Wales before March 1853
· p89: E.H. Corbould made a drawing for the 5/- design of New South Wales for £12 [issued 1861]
· p95-97: E.H. Corbould made various drawings, with one an essay for New South Wales showing “Queen in Coronation Chair” done some time before 11 February 1854
· p112-113: E.H. Corbould’s Chalon miniature for Van Diemen’s Land, May 1954
· p158: “No record of E.H. Corbould having been employed, nor was any charge made to the colony [New Zealand] for the drawing.”
· p336: The caption for E.H. Corbould's miniature for the Britannia design of Mauritius
· p408: E.H. Corbould made a drawing of the Queen’s head for a Ceylon issue, in April 1855
· p416: The caption for E.H. Corbould's water-colour drawing of the Queen’s head for a Ceylon issue
· p448: The caption for a die proof of the 4d design for Ceylon engraved by Jeens from a water-colour drawing by E.H. Corbould
· Plate XLVIII after p776: Reproduction of a handwritten receipt from E.H. Corbould.
· Plate XLIX after p784: Reproduction of a handwritten letter from E.H. Corbould (2 pages).
|
From this author’s perusal of the British literature, the claim that Corbould’s miniature influence the New Zealand Chalons seemed to have been confined to [Mottram1895] and [Melville16] and did not re-appear after 1953. Certainly when Robson Lowe [Lowe55, p345] considered what the engraver of the New Zealand Chalons used for a model in 1955, Lowe’s answer made no reference to Corbould or his miniature in this context whatsoever.
For R.J.G. Collins, as reported in [TPSoNZ64v4, p1-2], it was this de Worms reference that first triggered doubts about Corbould’s association with the New Zealand Chalons. Perhaps out of deference to [Melville16] or lingering confusion as to the nature of the Corbould miniature, he could not completely reverse his position, as evidenced by the following paragraph extracted from [TPSoNZ64v4, p3], “There can be no doubt that Chalon’s painting was the inspiration for … [the] engraving. As the Perkins, Bacon early records relating to bank-notes do not appear to have been preserved, the question of whether … [the engraver] used a Corbould water-colour sketch as a guide must remain controversial. However, in the circumstances association with the New Zealand stamps, the matter is only of academic interest.”
Nonetheless [TPSoNZ64v4, p3] (and earlier [TPSoNZ38v1, p32-33]) also describe the Perkins, Bacon & Co. process whereby an impression could be taken from an original die on to a transfer roller to yield a copy. [TPSoNZ64v4, p3] goes on to apply this information to the New Zealand Chalons: “The basic portrait was identical with the portrait in Humphrys’ original bank-note die. As was pointed out in Volume I, the expression of the face differs slightly in each value. This indicates that there was some touching up of the impression of the portrait ion each subsidiary die.” In this way, [TPSoNZ64v4] ties the New Zealand Chalon vignette back to the original bank-note die. Although no germane dates are cited, implicitly the origin date for the engraving of Victoria is pushed even further back than the 3 March 1854 date of the 1d uncleared essay.
In [Stone69n915, p65], Stone quantifies the date for the original bank-note die, and thereby moves the timeline even earlier: “In 1846 … [an engraver] made a three-quarter-length engraving of the Queen in her coronation robes after a portrait by A. E. Chalon, R.A., possibly for the … banknotes on which it was later used.” Stone goes on to comment on transfers, the Nova Scotia 1d die and thence a “Master Die [for New Zealand]. Engraved early 1854.” These comments are buttressed by [Stone69n911, p229-230] “Nova Scotia One Penny … The head was transferred from … [an] engraving of the Chalon portrait made for a … [bank-]note in 1846, and was extensively retouched. (For further details of this engraving see New Zealand hereafter)”.
Corbould’s miniature is tiny, at 19mm high [Bacon33] , and executed in water-colour. Apparently not expressed until 1991 in Smith’s vastly detailed work [Smith91], though recognisable in earlier times, the miniature water-colour could only provide limited assistance to the stamp engraver.
[Smith91] goes on to dispute that the Corbould miniature had any real impact at all on any stamp, let alone the New Zealand Chalons, and if anything the influence was in the reverse direction, writing “In May or June of 1854 … [an engraver] engraved the philatelic portrait head of the Queen for use of the Van Diemen’s Land one penny stamp. It is recorded that Perkins Bacon’s engravers were working on the die during the week beginning 10 June, and also that the firm had received on 16 May a small sketch by Henry Corbould showing a head and shoulders in an oval frame – i.e., as on the Canada 12 pence stamp. The sketch cannot really be considered as the source of the … [engraver’s] head, since it is not sufficiently detailed and the face itself bears only a passing resemblance to the philatelic engraving. … In summary, therefore, the writers’ hypothesis is that … [an engraving of the Chalon portrait] was the source for the ‘Chalon’ heads, and that the Corbould sketch was itself a copy – in principle only – of Jones’ 12 pence Canada stamp.”
The outstanding paper [Dickson2000] carefully reviews many claims around the Chalon stamps taken from a broad sweep of the historical literature. The paper captures mostly carefully what is known, and also what is unsubstantiated tradition. Still, Dickson leaves open the possibility that Corbould might have been involved one way or another via “Whether or not [Corbould] had anything to do with New Zealand stamp design …”
The exquisite [Odenweller2009, p8] hews somewhat closely to [TPSoNZ64v4] in regards to the influence of Corbould on the New Zealand Chalons. The book retraces the impact of [Mottram1895] and [TPSoNZ38v1] (as above) and urges caution with such speculation. Next the prior usage of the various elements (portrait, backgrounds) of the New Zealand Chalons is described, leading to “With the combination of such diverse elements, it would seem that any discussion of the influence of ... Corbould is not of much consequence.”
2. Rebutting [Melville16]’s claim
The argument against the Corbould miniature having any influence on the New Zealand Chalons has two legs.
First there is the absence of supporting information in the Perkins, Bacon & Co. records as discussed above [PBR53, Vol. I, p158].
Second, by the time that either of Corbould’s artworks existed, let alone reached Perkins, Bacon & Co (10 May 1854), the three-quarter length of Chalon engraving was already in existence (indeed, since 1846) and there were two partial dies for the New Zealand issue: the 1d essay (called the uncleared die in [Odenweller2009, p8]) of 3 March 1854 and the similarly uncleared 2d essay of 4 April 1854. Given these dates, Corbould’s artistry could have had no influence on the uncleared 1d and 2d dies.
But to complete the second leg it is also necessary to show that the (modest) changes from the uncleared 1d die to the issued stamps – characterized by such references as a 1d die proof, a 2d die proof and a 1d plate proof - are uncorrelated with the Corbould miniature water-colour, and instead the changes re-emphasise characteristics of earlier Perkins, Bacon & Co. engravings.
First we take a macroscopic perspective, and compare the uncleared 1d die with the 1d die proof. Since the 1d die proof seems over-inked (or the paper over-absorbent), we also compare the uncleared 1d die with a 2d die proof and a 1d plate proof. This seems fair since, if Corbould’s miniature water-colour influenced the 1d die proof, then the influence should immediately carry over to the 1d plate proof. Further the influence should also extend to the 2d design.
Methodology: Images were scanned at 1200dpi, rectified (since the reference work did not lie perfectly flat on the scanner), resized to exactly the same resolution, the colour histograms of the cleared images were tweaked to exactly match the unclear 1d essay image, then 128x128 pixel subimages of the cleared images were pushed around by a few pixels in order to get a more exact match (since at 1200dpi any uneven paper shrinkage etc leads to internal mismatches even if the borders are well registered). Then the two images, which should be almost identical (except for the value tablet for the 2d design, and any uncleared regions), are subtracted one from the other, and plotted. Since negative R or G or B intensity is not allowed, the difference is halved (to avoid overflow) and added to neutral gray.
 |
| 1d die proof with respect to the 1d uncleared die. Dark and light sections indicate that the 1d die proof was darker and lighter than the 1d uncleared die, respectively. |
 |
| Alternating views of the uncleared 1d die and the 1d die proof (animated gif). Copyright believed expired |
 |
| 2d die proof with respect to the 1d uncleared die. Dark and light sections indicate that the 1d die proof was darker and lighter than the 1d uncleared die, respectively. |
 |
| Alternating views of the uncleared 1d die and the 2d die proof (animated gif). Copyright believed expired |
We see that even this careful process is far from foolproof. Ink smudges in the 1d die proof appear as differences (at the 5 o’clock and 7’oclock positions of the outer circular border). There are pervasive, marked changes in the background yet they certainly were not triggered by Corbould’s miniature. They might be due to genuine touch-ups or imperfections in the in methodology described above, or some mix of both. Still, we can reasonably bound the differences to the shadow to the left of the nose, the shading at the Queen’s right check and neck and under the mouth, and also changes to the eyebrows, eyes, necklace and gown.
The second process was merely to concatenate heavily cropped portions of each pertinent reference work, and make an estimation: for each feature, was the end result closer to (or move closer to) an earlier reference (such as the smaller Cousins engraving, the three-quarter bank-note engraving or to the Corbould miniature? Due to this author’s artistic limitations, these questions only examine particular, low-level features rather than perform some holistic artistic comparison. As well, from the earlier discussion of the two Corbould printings, we use the Wilson image given its apparent lower level of “fuzz”.
The key to the concatenated images is shown below, where the light green boxes are the earlier references and the light blue positions are the end results:
Cousins smaller print © The Trustees of the British Museum.
Shared under a Creative Commons
Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) licence
|
184x Humphrys first Chalon banknote engraving [Humphrys184x]
Credit: The Library Company of Philadelphia |
185-/ 186-/ 187- Union Bank of Australia vignette (specimen)
|
1d uncleared die [PBR53, Vol. I, Plate XI]
|
Wilson print of Corbould miniature water-colour © Wilson
|
1d die proof [PBR53, Vol. I, Plate XI]
|
2d die proof [PBR53, Vol. I, Plate XI]
|
1d plate proof
|
The copyrighted images below have either been explicitly licensed, or their use is believed to be allowed because the purpose is research, the source works are published, the copies were obtained fairly, there is no competitive motive, and a minimal fraction of each work is reproduced.
Feature
|
Were the end results (1d and 2d die proofs and 1d plate proof) closer to the earlier references (Wagstaff, Humphrys' bank-note engraving, UBoA, uncleared 1d die) rather than the Corbould miniature?
| ||
| |||
Nose
|
Yes. The end results have a strengthened shadow to the left of the nose (as seen by the viewer), which spans the length of the nose like the earlier references yet unlike the Corbould miniature where the nose shadow tapers towards the tip of the nose.
| ||
| |||
Queen’s right eye
|
Yes. Most end results closely resemble the earlier references especially Humphrys' bank-note engraving, except that the iris in the 1d die proof is strengthened. Meanwhile the Corbould miniature has more of a side-on perspective leading to a more compact iris.
| ||
Queen’s right eyebrow
|
Uncorrelated. The end results have thickened the under-side of the outer end of the eyebrow. This seems to be a new characteristic, found in neither the earlier references nor the Corbould miniature. (The Corbould miniature is unclear, and there might be some thickening or bifurcation (!?) of the eyebrow but on the upper side, so this is a separate issue).
| ||
| |||
Queen’s left eye
|
Yes. Most end results closely resemble the earlier references especially Humphrys' bank-noteengraving , except that the iris and middle eyelashes in the 1d die proof are strengthened. Meanwhile the Corbould miniature has more of a side-on perspective leading to a more compact eye and iris, and the eyelashes are indiscernible.
| ||
Queen’s left eyebrow
|
Uncorrelated-to-Yes. Most end results do not depart too far from the earlier references especially Humphrys' bank-note engraving, except that the outer edge of the eyebrow in the 1d die proof is truncated before the outer edge of the eye. Meanwhile the Corbould miniature has more of a side-on perspective leading to a more arched eyebrow, and the edge of the eyebrow extends in line to the outer edge of the eye.
| ||
| |||
Under-mouth
|
Yes. The end results have a strengthened shadow under the mouth, which seems to be reasonably elongated left-to-right and reflective of Humphrys' bank-note engraving whereas the Corbould drawing, although indistinct here, seems to be rounder in shape.
| ||
| |||
Neck
|
Yes. The shadow on the right side of the Queen’s neck has two distinct regions in the earlier references: directly under the chin and over the shoulder, and these regions are separated by a lighter region on the neck. This is true for all designs except for the Corbould drawing where the shadow is continuous.
| ||
| |||
Earring
|
Yes. The earring in the Corbould miniature is very plain with no hint of pearls
| ||
| |||
Cross-Pattée
|
Uncorrelated-to-Yes. The diagonal slits in Humphrys' bank-note engraving cross-pattée are closed at the end, there is a central dome, and the bottom of the cross-pattée is a solid triangle. In the UBoA design and the uncleared 1d, two or three of the slits open up (and, for the top left slit, the UBoA design opens at the corner or 10:30 position but the uncleared 1d’s top left slit terminates at an 11 o’clock position). With the Corbould miniature, the diagonal slits are mostly open (and the top left slit opens at the 10:30 position), there is no central dome, and the bottom of the cross-pattée is weak. In the end results, all the slits are open (mostly at the 10:30 position but the 2d die proof opens at the 11 o’clock position), the central dome is more distinct and the bottom of the cross-pattée is a solid triangle.
Thus the changes to the slits in the end results are generally in-line with the Corbould miniature but also are in-line to changes in the early references too. Changes to the central dome and the triangular base in the end results resemble the early references much more than the Corbould miniature.
| ||
| |||
Necklace
|
Yes. The number of gemstones in the end results does not change relative to the earlier references, and that number is larger than the number of gemstones on the necklace in the Corbould miniature.
| ||
Score
|
Scoring: Ten features are analysed so each feature contributes 10%. This percentage is split three ways: is the feature A) correlated with the earlier (pre-Corbould) references, B) uncorrelated with both the earlier references and the Corbould miniature, or C) correlated with the Corbould miniature? These are written [A B C]. Thus a Yes in the above analysis contributes [10% 0% 0%], uncorrelated-to-yes contributes [5% 5% 0%], uncorrelated contributes [0% 10% 0%], uncorrelated-to-no contributes [0% 5% 5%] and no contributes [0% 0% 10%]. A score of [100% 0% 0%] would indicate that the end results are completely correlated with the earlier references rather than the Corbould miniature. On the other hand, a score of [0% 0% 100%] would indicate that the end results are completely correlated with the Corbould miniature rather than the earlier references.
Total scores:
80% of changes are correlated with the earlier (pre-Corbould) references.
20% of changes are uncorrelated with both the earlier references and the Corbould miniature.
0% of changes are correlated with the Corbould miniature.
| ||
3. Summary
It is this author’s argument that the engraver of the New Zealand Chalons had done the bulk of his work before Corbould’s initial sizing sketch or the miniature water-colour existed. Afterwards there were changes, but these changes were, in the main, highly correlated with restoring the characteristics of earlier engravings at Perkins, Bacon & Co. There is almost no example where a change causes the New Zealand Chalon design to look more like the Corbould portrait. Thus, if the Corbould water-colour had any influence at all, then it was the most minor in nature and really does not justify mention.
In some sense this conclusion is non-controversial since it accords with the author Percy de Worms’s comments when he states “There is no record of Corbould having been employed, nor was any charge made to the colony for the drawing.” [PBR53, Vol. I, p158].
Furthermore, with the benefit of hindsight, assisted by the cumulative literature, which is partially discoverable via RPSL’s web-based text search tool, and further assisted by image search tools that provide ready access to the artefacts of the Royal Collection and museums worldwide (i.e. with a grossly unfair level of assistance), this author proposes that, had the information known in 1938 been widely disseminated, the influence of the Corbould miniature on the New Zealand Chalons could have been strongly bounded:
· the literature demonstrated that the uncleared 1d essay preceded the Corbould miniature, and
· given the description of the uncleared dies in [TPSoNZ38v1], someone had evidently sighted them, and would be familiar with the issued New Zealand Chalons. That someone could have reported that the uncleared 1d essay very closely resembled the issued stamps except for such gross items as the essay not having the NEW ZEALAND nor the POSTAGE lettering cleared out.
No comments:
Post a Comment