My question is a little involved, so please bear with me.  Just from reading the description of the usage of -be’ in the Klingon Dictionary, I would assume that to negate the verb itself, as opposed to a specific meaning-bearing suffix, one would attach -be’ directly after the verb.

However, I did notice that Dr. Okrand apparently chose at least some of the sample sentences at the end of the book to highlight some of the subtler aspects of Klingon grammar (e.g., use of relative clauses with locations, constructing time expressions using subordinate clauses, etc.).  He also provides us with a couple of sample sentences that contradict my assumed rule stated above.  For example, vIta’pu’be’ and pa’ jIHpu’be’.  In both of those sentences, -be’ appears after a type 7 suffix.

I haven’t really studied agglutinative languages in the past, so I’m not sure if there are some real world analogues that could shed some light on this, and I don’t have access to a substantial Klingon corpus to work through looking for clues.  So I have two working theories, and I am hoping someone here could tell me whether either (or neither) theory is correct:

Theory 1: Aspect suffixes are more intrinsically linked to the essential meaning of the verb (in contrast to, say, type 2 or type 6 suffixes that act like English modal verbs or add other information descriptive of the verb). Therefore, -be’ will follow an aspect suffix since it’s still coming after an intrinsic element of the verb, and -be’ should come after the last such suffix.  If this theory is true, then the question becomes which other suffixes also have more intrinsic meanings.  Certainly, an argument could be made that type 1 and type 3 suffixes, along with -lu’, have more intrinsic meanings.  But what about, say, -moH.  I could see arguments both ways.  So should the sentence “I do not clean it” be translated as vISay’moHbe’ or vISay’be’moH? 

Similarly, let’s say I instructed a third party to prevent someone else from eavesdropping on a conversation, and then I reported “He will not let them hear.”  Would that be Qoybe’moH or QoymoHbe’?  (Although again, you can probably parse those two sentences differently with slightly different meanings.  “He will ensure they do not hear” vs. “He will not cause them to hear.”)

Theory 2: Similar to -Qo’, -be’ wants to occur as far back in the stack of suffixes as it can get.  So, under this theory, -be’ will attach after the very last non-type 9 suffix unless that would change the meaning of what I’m trying to negate.  So, for example,  vIleghnISbe’ means something different from vIleghbe’nIS (“I do not need to see it” vs. “I need to not see it”).  In a situation like that, -be’ comes after the element that makes the most logical sense.  But in a situation where the substantive meaning wouldn’t necessarily change by placement (vIta’be’pu’ vs. vIta’pu’be’), the -be’ comes in as late as possible.

Hopefully I’ve made enough sense that it’s clear what I’m asking.  Any advice/guidance/assistance would be most appreciated!