Proposal: Redefine attitudinal default scope as ambiguous

Initially proposed by la zipcpi.

The status quo

CLL 13.2 states that attitudinals and UI-cmavo always bind to the previous lexical unit. For example, la .djan. cu klama iu (cu now required by current grammar) specifically means that it is the coming (lo ka klama) that inspires the "love" emotion of iu, and not the entire sentence (lo du'u la .djan. cu klama). The current cleanest way of attaching UI-cmavo to the entire sentence is to put them first, e.g. [.i] iu la .djan. cu klama. The other way is to close the sentence with the appropriate famyma'o (la .djan. cu klama vau iu). (If the last word is either the starting boundary (e.g. .i, lo, cu) or the ending boundary (e.g. ku, kei, vau) of a lexical unit, the UI-clause is considered to affect the entire unit.)

The problem

Attitudinals are often added as afterthoughts; in online chat systems in English and other languages, smileys are often attached to the end of sentences, rather than the beginning. It is often undesirable to announce "I am going to tell a joke" (zo'o) or "the following is sarcastic" (xo'o) before making a joke or sarcastic comment. Thus the "correct" prescribed way of doing that is to use famyma'o like vau to close sentences before adding the attitudinals. However, this is often forgotten in usage, and vau is sometimes insufficient to close sentences completely, and is sometimes even ungrammatical. Additionally, famyma'o are a difficult concept to teach to beginning learners of Lojban. I go more in-depth on the problems with this on my zo i'au ki'a page.

The proposal

Attitudinals and other UI-cmavo (like xu) are now considered by default to be inherently ambiguous as to scope. It could bind to the last lexical unit, but could also apply to any number of units before that; in other words, all attitudinal groups have an implied fu'o, and its corresponding fu'e would be placed by context.

To allow the speaker to resolve this ambiguity when necessary, two new cmavo are to be proposed, in the new sub-selma'o UI8 ("attitudinal scope modifiers"). i'au, as in my previous proposal, binds an attitudinal group to the sentence. (Under this proposal, I expect it to be used a lot less often than under my previous proposal, as the usual contextual ambiguity-resolution to afterthought attitudinals is, generally, "the sentence or statement". i'au might still find occasional use when it is important to explicitly define the scope.) ji'au is a new cmavo that will bind an attitudinal group to the last lexical unit, as UI-cmavo are originally defined to do. So now to specifically state that it is lo ka klama that inspires the iu emotion, you say la .djan. cu klama ji'au iu

Expected advantages

• This proposal better-respects current usage, as people often forget to close their sentences, or close them improperly, before adding afterthought attitudinals. Even ro lo ma'a datka has this problem (is lo ka rebla the inspiration for ua?), and adding the appropriate vau or i'au would completely ruin the meter of the song. (Actually that might not even work, since the last sentence is just .ije galtu rebla, while ua probably applies to the entire situation of the song verse.) Because of how common this "incorrect" usage is, it is now hard to explicitly apply an attitudinal group to the last lexical unit of a sentence in the occasions it is necessary; under this proposal, one would now use ji'au in those situations.
• There has been some opposition against my push to make i'au the prescribed way of attaching afterthought attitudinals to Lojbanists of all levels, and especially against my push to make it monosyllabic to better serve this purpose. This proposal, by making the "default" scope being dependent on context, will make it unnecessary for any new cmavo to be so strongly prescribed for general conversational use; i'au and ji'au will now only be used when it is important to explicitly define the scope of an attitudinal or UI-cmavo group.