How do clusive forms arise?

Question

Most non-European languages exhibit a clusivity (exclusive/inclusive) distinction. What are the common ways of developing new clusive forms and which clusivity is tied more tightly to the first person singular?

In Washo (an Amerindian language), the inclusive dual -ši and plural -hu are the marked forms, marked by a suffix: this suffix is the same as the non-first-person-restricted dual and plural suffixes for independent pronouns. The exclusive is not marked at all.

In Miskito, however, the first person inclusive is formed from first person "yang" and the plural marker "nani" to form "yang nani", while the first person exclusive is marked: "yawan" (no singular counterpart) without an explicit plural marker.

So one marks the inclusive, the other the exclusive.

Answer 1

Link between clusive pronouns and first person singular

The Wikipedia article for clusivity does include a somewhat incomplete table of the inclusive/exclusive pronouns in various languages, including a column listing what form is closest to / related to the first person singular. As far as I can tell, the data seems fairly mixed, and there are no individual citations for these forms.

For a more theoretical perspective, which might shed light the relationship between clusive pronouns and other pronouns, the one JSTOR article that I could find which might be relevant was:

On Markedness Asymmetries in Person and Number Martha McGinnis Language Vol. 81, No. 3 (Sep., 2005), pp. 699-718 Published by: Linguistic Society of America Article Stable URL: http://www.jstor.org/stable/4489970

I perused this article. It discusses a revision of a feature representation of pronouns proposed by Harley and Ritter, and some markedness asymmetries. I.e. if dual number is represented as [+group] and [+minimum], or some sort of combination of the singular and plural, why do languages with no distinct dual category always conflate the dual with the plural, and never the singular? Similarly, in languages without an inclusive category, the inclusive is always conflated with the first person, not the second.

I am not sure if this will be of any value to you; it does not answer your empirical questions directly, but does discuss cross linguistic evidence.

Development of Clusive Pronouns

Another paper by Osada Toshiki on historical development of clusive pronouns in South Asian languages can be found at: http://www.sealang.net/archives/mks/pdf/34:79-96.pdf. It discusses several different possibilities for how languages could develop this distinction, including areal diffusion and internal development. Areal diffusion seems to be far more common.

However there are plausibly cases for internal developments of clusive pronouns. For example, Japanese has a special exclusive pronoun temae-domo (watashi-tachi or watakushi-tachi being the general first person plural). This is analyzed as temae 'I' (further decomposed as te 'hand' and mae 'front') and domo (plural suffix). It's only used by sales persons, presumably motivated by the semantic need to differentiate a group including the sales people and customers and a group only of sales people. The article rejects an analysis of the feature being introduced by contact with Ainu because that would not explain its limited use (only by sales people). Also temae is not attested before the 16th century, too late for Ainu influence.

Another case of internal development is possibly attested in Indo-Aryan languages, in which a reflexive pronoun became a second singular honorific pronoun which in turn became a first person inclusive pronoun.

Best of luck.

Answer 2

Linguistic description of Berber languages does not mention the clusivity as phenomenon that exists in Berber. However, according to me, such particularity is present in Riffian at least for the inclusive. This form is generally considered a imperative form, but it is not the case.

Inclusive "we" is only used in the future tense and is constructed like this:

a n-ecc-et = you and me are going to eat

a(d) = Future

n- .... -t = we inclusive

ecc = eat

Here, we inclusive is a circumfix that is formed with n- and -t which are morphemes coming from other paradigms:

a n-ecc

Fut we-eat

We will eat/ are going to eat

ecc-et

eat-you.plur.IMP

eat

As we can see, we inclusive is the combination of two different grammatical persons (first-person plural and second-person plural) and two different moods (indicative and imperative)