Evaluating armorials (II) – Variant coats of arms
The first part of these series on evaluating armorials dealt with the identification of coats of arms in armorials. It appeared that many armorials were copied and that copyists were prone to make mistakes. This sometimes leaves the researcher with different coats of arms for the same family. This post deals with the problematic of these variant coats of arms more thoroughly by discussing their origins and how they should be analysed. A lot of heraldic terms are used in this post. In order to make this article intelligible for everybody, we have created a short dictionary of the terms used by Steen Clemmensen.
Coats of arms belonging to the same family may vary in their depiction between sources. Some variations (tincture of crowns, beaks, claws etc.) are of little or no practical importance when studying armorials or the bearing of arms. Other variations, like the absence or presence of crowns, tail forms, heads facing guardant or not, and differences in the number of secondary figures may be important for discussions of the usage of arms by branches, but not for comparing segments or individual arms in armorials. Some representations varied with time, e.g. posture of the lions in the royal arms of England or Denmark. In any case, there are too many examples of such variations in the rendering of arms in armorials from any period or country. If readers disagree, please peruse the various books and manuscripts.
Evaluation of variants gets a little more complicated when elements or brisures appear or disappear between manuscripts, or when single arms are replaced with quartered or quarters replaced in composite arms. Such differences in single arms or individual quarters can often be ascribed to sloppy copying. Replacement of quarters or exchanges of single arms, e.g. for a finer lordship or title, would indicate an update preferred by the copyist or compiler. The key question is: which variations are important for what?
Different kinds of variation
In essence, variant arms come in two kinds:
- involuntary mistakes in rendering or copying a coat of arms, which can be classified according to type and seriousness; and
- variant forms actually used by a branch or a person.
For the latter, the researcher must determine whether the number or colour of a minor brisure, e.g. one or more martlets, would be regarded as important. A single example may suffice. For the French family group related to the Châtillon-sur-Marne at least 133 seals and 266 items from 50 armorials are recorded with altogether 77 different blazons, all based on Gules 3 pales vair and a chief or. Depending on the criteria used, one may divide the family group into some 30 armigerous branches, not counting a few individuals with personal brisures. For armorials some of these variants are easy to account for. One (probable) sideline, the Montchâlons, have the field vert. Other branches have a distinct cadency mark, such as a lion issant or a martlet. Changing from one cadency mark to another would belong to the involuntary mistakes, though one or more members of a branch could choose to use a different brisure. Changes, abbreviations or illegibility of legends make it difficult to place a set of items within a certain branch, but that is a different story. The major problem concerns the 72 items with 11 different blazons having 1-4 martlets (merlette in french) or mullets (molettes) in either gules, azure or sable. Some of these are undoubtedly real differences, but others are not – but which? The possibilities for mistakes in copying include misreading the blazon (but hardly the visual figure), substituting a different minor brisure (the family also used labels, lions and fleurs-de-lis), or using the wrong paint on a small figure.
Classification of variants
For the discussion of which variations might be important, the following eight different kinds of variations, categorized into three types, may do for a start:
|Type I: These variations are serious, as they render the shield in a completely different way|
|Type II: These variations are somewhat troublesome, but can relatively easily be reconciled.|
|Type III: These variations are insignificant for comparing or using armorials. They are very numerous, and may be attributed to sloppiness, change of fashion or simply the preference of the artisan or armorist rendering the figure. The choice of blazoning a coat of arms as per fess or with/on a chief can be regarded as a special case of no. 8 (partial change) or even no. 4 (wobbling) due to local fashion. The Anglo-French preferred chief for most situations|
The occurrence of different blazons (type I, no. 1) is problematic. This might be due to differentiation by branches – and be attributed to another branch, if the owner in question is not known to the commissioner, reporter or artisan. It could also be due to misreading a damaged blazon or painted shield, or simply forgetting to add a brisure or secondary figure. Some figures are hard to differentiate: fox, wolf or marten could be turned into lions, or a bear’s head into a talbot’s or lion’s head. The responsibility for inversions (type I, no. 2), and probably also colour changes such as Gules-Azure, can probably be attributed to the artisan or reporter. Everyone who has worked on transcribing or copying long lists of arms will know the embarrassment of finding this kind of mistakes when comparing his copy with the source. Similar mistakes would also be possible for an observer at a tournament or any other gathering.
Of the type II variations, reversions and flipping may reflect the layout of the source, or how the arms were faced when noted originally. If a coat of arms was first noted in a church, it would normally have faced the altar, i.e. the head of an animal would be turned sinister if the arms was hung on the north wall. Similarly arms correctly Party per bend would appear as Per bend sinister, and Quarters 1+3 (for the viewer left) would look like Quarters 2+4 (for the viewer right). Some armorials have the shields facing the center of the page, e.g. parts of the Rugen, which gives the same result for the leftmost columns. Unfortunately, the artisans were rarely consistent, so some arms appear as they were borne, while other appear as mirror images. But quarters might also be reversed due to the importance placed on the first quarter, which might reflect the more important lordship or family – as seen by the artisan or reporter. It might even be the owner himself who has changed the order. Flipping may also be due to mistakes by an artisan, either in colouring or misreading his source. Wobbling might equally be due to changing fashions, e.g. like the chevronny, which was later replaced by three chevrons, or to misreading or misdrawing, e.g. paly of 8 turned into 3 pales. Fading is self-evident. Most copyists have experienced taking the wrong option when transcribing a manuscript – and as for fading many copies were made when the source was already old.
Most of the type III variations concern animals, which are normally crowned and armed according to a standard colouring scheme: crowns Gules for body Or/Argent, Or for other colours; langued Gules, but Azure for body Gules; and armed Gules, except Or for body Gules or Azure. Artisans may vary this scheme, when the field has one of the colours of the crown etc. Most deviations from this standard scheme can be explained as the artisan using the same brush for colouring several secondary figures on a page. Crowns and posture were clearly significant elements in any coat of arms in the Late Middle Ages, and ought to be considered as an important change when evaluating a family’s use of arms. However, looking across several armorials one can find both lion, lion crowned, lion guardant and lion guardant crowned for the same family, e.g. End von Grimenstein in 11 armorials. Posture (passant, statant, salient / rampant) may vary according to the number of figures and their place in the shield or quarter. Whether to count such changes as important or serious depends on the setting. It may be unimportant to have one animal changing posture, but to have 3 lions passant all in pale change into 3 lions rampant (2:1) will only happen, when a blazon was seriously damaged before it was copied.
The next post in this series will deal with the practise of comparing armorials, for which the variants and mistakes that have been discussed in the last two posts can be quite useful.
 For this discussion, family is the short term for a combination of arms and name, which besides the family proper may represent a branch using a specific combination or a diocese, abbey, town, office, guild or country.
 Steen Clemmensen: The use of arms by the Châtillon-sur-Marne during the Middle Ages. Farum 2014, www.armorial.dk .
 The same courtesy effect is present in the 1605 version of the Siebmacher with items printed in rows of five. The outer pairs are turned towards each other, with the center columns facing the reader – easily recognized from the posture of the helmets. Many readers do not realize this, when taking the arms down in blazon or drawing – getting not the actual arms, but its mirror image, flipped or reversed.