Scalar Implicatures across Languages?

This weeks selected readings served as an introduction to implicatures and, more specifically, scalar implicatures, phenomena whereby weaker linguistic terms by virtue of their being chosen over stronger alternatives of the same scale (“same” versus “all) elicit a more precise semantic meaning. Variables thought to influence ability in comprehending these implications include the following: processing memory (contested/debunked), development of scales, and contextual information among others.

Because we are interested in how this ability develops, we look to its manifestation in children and find that it does not inherently exist. It is learned to some extent. Children have to build the proper scalar relationships, similar to building a vocabulary, before they can apply this impliciative ability, which seems to be more based in logical reasoning than in Language.

From what I understand, and what is shown in the study using contextual implicatures in its methodology, the ability to correctly interpret these implications is innate. It is the ability’s linguistic manifestation that needs to be developed.

What I would like to know is how this ability presents itself across different languages and in different communities. In particular, what are the distributions of scale density (number of logical alternatives defining a qualitatively equivalent range) among differently parameterized languages? Additionally, it would be interesting to see to what extent exactly choosing any scalar quantifier over another, paralleling the effect seen in quality rarity, which it is suggested influences how much information a description expresses, is present in these linguistic scales across languages. Could saying “some” versus “all” evoke a different experimental distribution (given the same experimental procedures described in the readings) than the equivalent words in other languages?

One question that this phenomenon evokes for me is such: What is the best way to interpret these implications, and where is the overlap between this interpretation and how they are naturally understood?

One distinction which both readings made clear was that the trends in the data were established for complicit or cooperating speakers. This label is important. It only accounts for speakers who are not aware of and have no intention to manipulate a listener’s interpretation of the implication in their speech, from scalar implicatures or otherwise. It would be interesting to see if there exist any legal precedents for the interpretation of these technically ambiguous, superficially explicit statements. Where does being a “non-cooperating” speaker fit in with the law?

On one final note, although the data collected was statistically significant in some cases, what would lead to such a split in interpretations across adults given the task of rejecting a certain description; that is, why do so many accept “some” in cases where the majority requires “all?”