Saltar para: Post [1], Pesquisa e Arquivos [2]

St Edward the Confessor: Shield's Redundancy (6/6)

Published by Carlos da Fonte, em 11.09.12
Edward the Confessor - Attributed Arms

This sixth and last semantic level was so inconspicuous that we've just found it at the very end of our analysis. One of the reasons was that it doesn't categorically affect any heraldic traces as colours or shapes and works as a redundancy for everything already treated. For this level we used the epithet of Edward  - the Confessor - which we may classify as an anthroponymic form of the corresponding referent's metonymy. It is interesting to observe that compared to the other descriptions we've used: King, Saint, Edward and Wintonian, this last is the most characteristic. The other four only make sense when together but “the Confessor” may evoke much faster our referent Edward when alone, for many kings, saints, Edwards and Wintonians existed.


As there was no heraldic effect resulting from this level we should doubt the adequacy of considering it in the analysis. A good advantage is to screen any incompatibilities among other levels. This, of course, would be much more favourable if detected during the initial analytical stages. The most difficult part of our studies is to achieve proper parophonies that fit well one to another inside a heraldic plot. In the beginning of every investigation, when we know nothing about the plot used, once a tendency is found, all the semantic levels must obey it.


Another benefit is to ascertain if such behaviour reproduces consistently in our corpus and eventually to discover other associations that will only be apparent by observing all occurrences as a whole. That's how we were able to identify most of the structures implicit in parophonies and later this helped us to develop several typologies in order to assist the establishment of relationships within distinct coats of arms.


The motivational scheme is quite peculiar in this example of St Edward's arms as the plot seems to have two different authors, set apart by hundreds of years. The early part, connected with the numismatic representation, refers Edward as a king; the late part depicts his glorification as a saint. If we didn't know the existence of the coins things would be much more difficult. The semantic examination may have diverted to other, presumably worse, solutions or simply stall because, for example, the most frequent Anglo-Norman word for king is rei and not roi.


The parophony we've found is Confessur (ano. Confessor) ~ Qu'hom fait sur (ano. that we make on). The modern form in French, qu'on, hides the origin of the word, linked with the impersonal hom (ano. man), acting as “someone” or “we”. The phrase is incomplete and we must look after the word that would accomplish a possible meaning. In the present context we may only find the substrate of the blazon as such a word. After deciding on the figurations and tinctures to be used all we need to do is draw and paint them on top of the shield. That's the subject that qu'om fait sur justifies.


The discretion index totals k = 0.14 and the average of the six levels is k = 0.08, an extremely low result, apparently indicating that most if not all proposed solutions cannot be improved regarding parophony. Note that this methodology doesn't supply proofs for any specific parophonic hypotheses we make. Only the whole proposition with its semantic levels, may be eventually regarded as coherent, therefore almost statistically impossible to result from chance. But even then we cannot guarantee that one or two of these levels aren't wrong. Finding other arms that repeat the same kind of association and visual behaviour is important for a sound justification, however the best would be to use plain historical evidence as documents and artefacts. This is not always easy and virtually impossible for attributed arms, as these we've finished studying now.


Note that we took the phonemes [Om] paired together with [Õ] which may look strange. In fact [O] and [m] are two different phonemes but they have necessarily to compare with a nasalized [Õ], which is a simple sound. If we followed the methodology blindly and used the formal pairing [Õ][_] ~ [O][m] the penalties used in the calculation would be excessively high and their numerical effect in the discretion index the same as, say,  [Õ][_] ~ [O][k], which is unacceptable. Note that this same criterion was applied before with [tS] and [S] for the parophony Itchen ~ I chenne.


We finished thus this analysis, one of the hardest we ever made with a record number of six semantic levels. That's the same number found in the primitive coat of arms of the kings of Portugal, which took most of our effort and time. Is it all true? We don't know, but remembering that all discretion indexes are quite low and that semantic levels interact coherently the answer will tend to be positive.


Let's introduce some ideas on a probabilistic proof. Take the parophony Seint  ~ Cinq as an example, and try to find other puns with Cinq or equivalents as V, B, quintet, etc. and pair them with other Old French words related in some consistent way with St Edward. To be comparable with our analysis the discretion index couldn't be higher than 0.2. For simplicity we're using a plain parophony with just one word. I guarantee that you won't find that many but suppose by exaggeration that you are able to discover ten viable parophonies. Then you must divide ten by the number of different words that existed in Old French which were able to be employed by an average speaker, let's say five thousand.


Then, also in a simplified manner, the probability that such a parophonic solution could result from luck is about 10/5,000 = 0.002 or 0.2% (with hypothetical equiprobable outcomes) .  If you extend this result to six different and independent semantic levels the odds for a simultaneous coincidence would be 0.002 × 0.002 × 0.002 × 0.002 × 0.002 × 0.002 = 0,000000000000000064 = 0,0000000000000064 %. Of course if you take composite parophonies, using more than one word, you must take into account all the combinations possible, two by two, three by three, etc. The result would be even smaller because the divisor is much bigger. Naturally, this is just an abridged approach that takes ideal and less complicated elements. But it's easy to understand and gives a reasonable picture on the orders of magnitude involved.



 Edward the Confessor - All
Attributed Arms R Edward the Confessor
Anthroponym M Confessor
Language of Conquest V Anglo-Norman
Denominant A Confessur
Graphemization A  C  |  O  |  N  |  F  |  E  |  S  |  S  |  U  |  R 
Phonemization A k  |  Õ  |  f  |  E  |  s  |  y  |  R\
Pairing A k  |  Õ  |  f  |  E  |  s  |  y  |  R\
A k  |Om|  f  |  E  |  s  |  y  |  R\
Coefficient of transposition A 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0
Coefficient of character A 0.0 | 0.5 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0
Coefficient of position A 0.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0
Addends A 0.0 | 0.5 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0
Discretion index A k = 0.14
Phonemization A k | Om | f | E | s | y | R\
Graphemization A Q| U' | H | O | M | | F | A | I | T | | S | U | R
Designant A qu'hom fait sur
Comprehensive E qu'hom fait sur
Simple monosemy, Redundancy S that we make on (the shield)
S qu'hom fait sur
Tincture H Azure
Number H a
Figuration H cross
Aspect H flory
Placement H cantoned with
Number H four
Figuration H martlets
Connective H and
Number H another
Placement H in base
Tincture H or


(next analysis in this blog is here)

Autoria e outros dados (tags, etc)

Published at 17:38

All rights reserved
© Carlos Carvalho da Fonte 2009-2017




  Pesquisar no Blog

subscrever feeds


Setembro 2012