Print

Print


Nobody has commented on this topic, I keep commenting myself. :-)

I am wondering if the following features of two of the different
algorithms described could be combined to simulate the effect of
government on language shift:

> * language "fitness" defined as the sum of the capabilities (natural resources) of all
> sites that language occupies;
[...]
> * with a certain probability, a person discards the mother language
> and adopts the language of another person.

I imagined something like a state controlling the access to natural
(and artifitial) resources on an area, so stimulating people on that
area to shift their language to the state's official language.

I've noted that many people speak English even here in Paris, either
at the university with foreign students or in turistic zones.
Sometimes some of them start speaking English to me when they note I'm
a foreigner, even when I started the conversation in French. So I
wondered if this type of thing could slowly take the whole world to
shift language towards English. I came to the conclusion that what
avoids this shift in many countries is that English (or whatever
"cultured-people" language) is not the official language, so the local
language keeps being more useful. It's not the case of some Amerindian
students I have at the university that must know Portuguese to access
the classes' information.

BTW, English seems to be the new "non-Barbarian language" as Greek and
Latin were before. Every cultured person is expected to know English.

Até mais!

Leonardo


2013/1/30 Leonardo Castro <[log in to unmask]>:
> I would like to hear some comments on how realistic do you think that
> the assumptions of computational simulation of language spread are:
>
> -> Viviane model [1]:
>
> * language simply labelled by an integer number;
> * heterogeneous environment with square cells of different
> "capabilities" (related to natural resources) homogeneous distributed
> between 0 and 1;
> * language "fitness" defined as the sum of the capabilities of all
> sites that language occupies;
> * probability of occupying a new neighbour site directly proportional
> to the language fitness and to new site capability;
> * probablility of language mutation (creating a new language labelled
> with another unused integer) inversely proportional to its fitness.
>
> -> Modified Vivane model [2]:
>
> * each language is described as a 16-bits string;
> * when a language occupies a new site, one of its bits can change;
> * if the bit strings of the languages of two sites are the same, they
> are the same language (so I presume that the original language can be
> reobtained from one of its mutated versions);
> * as "there are more bad places than good ones for people to live",
> the capabities C of the sites are distributed inhomogeneously with
> frequency proportional to 1/C;
> * in each time step, they choose two sites and occupie that of higher
> capability C.
>
> -> Schulze model [3]:
>
> * simulation of N people initially speaking the same language (or grammar);
> * each language (grammar) is described as having F features with Q
> possible values each one;
> * each feature changes with probability p in each iteraction;
> * with probability q the mutation above is not random, but a "copy" of
> the value of a neighbour person;
> * with probability (1-xˆ2)*r, a person discards the mother language
> and adopts the language of another person.
>
> ---
>
> [1] http://www.sciencedirect.com/science/journal/03784371/361/1 (article # 30)
> [2] http://arxiv.org/abs/physics/0608204
> [3] http://arxiv.org/abs/0704.0691
>
> Até mais!
>
> Leonardo