Codasyl VMS word grammar

Idea 9 July, text, coding and debugging 12‐16 July 2023

Introduction

As promised and announced, here I’ll try to recreate parts of Jorge Stolfi’s VMS word grammar in a different form, inspired by legacy Codasyl grammars for Cobol and Network Databases.

Word structure

CrustPrefix:
` [[[o],[y]]q] [[[o],[a],[y]]…{[d],[l],[r],[s],[n],[x]}]… `

MantlePrefix, Core, and MantleSuffix:
` [ [[o],[y]] {[ch],[sh]} ] [[ee],[oee]] [[e],[oe]] [[o],[y]] [c] {[t],[p],[k],[f]} [h] [[e],[oe]] [ [[o],[y]] {[ch],[sh]} ] [[e],[oe]][e] `

CrustSuffix:
` [[[o],[a],[y]]…{[d],[l],[r],[s],[n],[x]}]… [ [[o],[a],[y]] { [[[o],[y]]], [[[a],[o]]m], [[[a],[o]]i…[[n],[r],[l],[m],[s]]] } ] `

Comments

I must admit I cheated quite a bit here: my description is not quite complete, and not exactly the same as that by Jorge Stolfi. Notably:

• Gallows is either G or c.G.h, with G being one of the gallows letters t, p, k or f. However by making both [c] and [h] optional, my version would also allow c.G and G.h, which is different.
• I simplified the definition of MtS (Mantle Suffix) considerably, by removing quite a lot of the somewhat more infrequent variants.
• I ignored WholeCrust and WholeMantle, which is not fair.
• And there are probably more differences.

However, I think all of that that doesn’t invalidate my point, which is that a monk – or a more modern forger as well – with the help of a schedule like the one I just showed, could generate plausible looking but not necessarily meaningful words, which have a com­pli­cated structure with prefixes and suffixes, like in natural languages of certain typologies. Fusional? Agglutinative? Polysynthetic? Oligosynthetic?

To cover also the WholeCrust and WholeMantle cases that Jorge Stolfi identified, more than one such schedules would be needed, offering extra initial choices for the scribe or scribes.

The generation might have been done while writing, applying the several options each time, or perhaps a large list of frequent words might have been compiled first, and then used while writing. Or both methods combined.

Any way it was done, it is a hell of a lot of work, but I think it IS feasible, given enough time, motivation, perseverance and a mind set for this type of work. Some people do have all of that. Some people just are like that.

The accuracy that this requires, can exist side-by-side with a looseness, forgiveness and tolerance that can explain the large list of words found, by Jorge Stolfi and John Grove, that do not exactly abide by the rules.