On Mon, 18 Mar 2019 at 18:24, Felix Malmenbeck <felixm@kth.se> wrote:

On Mon, 18 Mar 2019 at 17:49, Lieven L. Litaer <levinius@gmx.de> wrote:

> The use of the suffix {-Ha'} does not imly that the situation or the
> action was different before. It's just the opposite meaning.

While there are some canonical examples that may suggest this to be the case (words like {jaQHa'} come to mind), the original description in TKD suggests that it requires either an undoing of a previous state/action, or that something is done wrongly:

"This negative suffix implies not merely that something is not done (as does -be'), but that there is a change of state: something that was previously done is now undone. For convenience, it will here be translated as "undo", but it is closer to the English prefixes mis-, de-, dis- (as in "misunderstand", "demystify", "disentangle"). It is also used if somethign is done wrongly. Unlie -be', -Ha' can be used in imperatives."

Having gone through every canon instance of {-Ha'} I could find, I'm now convinced that the original description in TKD is wrong. Either it's very incomplete, or Okrand changed his mind. {-Ha'} seems to indicate the opposite meaning (opposite state or action) most of the time, and only rarely an undoing or an error.

TKD 4.4 also says that {-qu'} is the only verb suffix possible on a verb which is acting adjectivally (by following a noun), which means {-Ha'} isn't allowed. However, canon contradicts this with {SuvwI' quvHa'} (HQ 12:3) and {Duj ngaDHa'} (KGT). (Incidentally, {-be'} appears to be possible too, as we also have {wa'maH yIHmey lI'be'} (PK). So the rule isn't that only {-qu'} is allowed, but that only rovers are allowed.)

De'vID