Sequential word repetitions in the VMS

[Anton]

I decided to have a recap of what has been proposed in this thread over eight years and here's my summary:

1) The sequential repetitions are just underlay-language-natural, becuase many phenomena can cause it, such as:
- syllabic/homophonic (like "pur Purpur" etc)
- poetic/incantation/invocation (like "kyrieleis kyrieleis kyrieleis" or "rise! rise! rise!")
- emphasis
- in some languages (like Chinese) sequential repetition is just normal beyond any special circumstances

However, the whole group of these explanations does not agree very well with the (high) density of repetitions, in the first place, and with the (large) amount of distinct vords subject to sequential repetition, in the second place.

2) The recent suggestion of numbers

Does not agree very well with the large amount of distinct vords subject to sequential repetition. Also, triple and quadruple repetitions imply operating with large numbers - in the region of hundreds and even thousands.

3) The effect of uncertain spacing

4) Scribal errors - dittography or haplography

In fact, the simplest explanation. With the large amount of quasi-reduplications it is likely that some of those would transform into exact reduplications through scribal errors. However, this implies that the whole text has been blindly copied from another source.

5) Side-effect of some vord-level transformation process, like shuffling/transposition. This still lacks a quantitative estimation of how probable it is that a text without reduplications would transform into a text with plenty of reduplications through the said process. Besides, I checked with one if the balneo folios (don't remember the number offhand, but that was that qockeedy-heavy folio with the quad-reduplication), and transposition of the text did not remove all reduplications. At least one reduplication remained.

6) Autocopying to fill blanks (left from some previous encoding process). I consider this worth of further investigation, especially given that, as pfeaster wrote above, reduplications tend to avoid line-start and line-end positions. It would be useful to explore if there is correlation between reduplications and baseline jumps.

**

I would also like to throw in another idea which recently occured to me, and that is: the ordered listing.

Imagine two sets which have a many-to-one relation between them. For example: one set is the set of the days of week, like this:

{Monday, Tuesday, Wednesday, Thursday, Friday, Saturday, Sunday}

The other set is the set of meals, like this:

{Meat, Fish, Vegetable}.

And the relation is established like this:

Monday -> Meat
Tuesday -> Fish
Wednesday -> Fish
Thursday -> Vegetable
Friday -> Vegetable
Saturday -> Meat
Sunday -> Meat

Then, describing this relation in text, you would obtain three double repetitions or two double and one triple repetition, depending on your starting point, like this:

The sequence of meals for days from Monday through Sunday is as follows: Meat, Fish, Fish, Vegetable, Vegetable, Meat, Meat

or

The sequence of meals for days from Saturday through Friday is as follows: Meat, Meat, Meat, Fish, Fish, Vegetable, Vegetable

This explanation faces the same problem as the clause 1) above - there would be in excess of 100 notions to be included in ordered lists, but otherwise it's quite elegant, I think.