Michael Sean Quinn*
The law on the topic under discussion here is almost the same everywhere, but not completely. As is common in Quinn’s blogs, however, Texas law is used as a paradigm, for better or for worse. Much of what is said here is easily adaptable to the Federal Rules. 

One thing that I believe to be true for all the systems is that they are all poorly drafted. Their purpose and their language do not match up.  

party must be cautious about writing “Admit” (“A”) as a response, given the dramatic
consequences it has on or for trial and summary judgment. It is taken to be conclusively
established that if proposition
p is
admitted by a party, then
p (without
a court decision) is established in that litigation. This means that it is established as “true”

Exactly opposite cannot be
true for a denial (“D”). If that were true, then if the Respondent denies a
proposition then the contradictory of that proposition would be conclusively established, as so would have to be thought of as true.  
 Obviously, when admitting the truth of a
proposition the party
 responding  needs to be quite sure, legitimately—indeed he must be quite certain, since there is little—though some–chance to turn

So, this is not the way a D is considered. The denial of a proposition presented in discovery does not entail the contradictory of the proposition denied. A denial is more like “I doubt it, so prove it.” This too must have a
legitimate foundation. One cannot deny that the sun rises in the east, simply
one tends to be a skeptic, even though one has the right to be skeptical.

proposition’s being known to a person means that the person has a relevant
belief which is both true and justified by evidence. If a proposition is not
known to a party, a denial of the proposition is (or may be) appropriate in the
context of requests for admission. A denial of p does not require that the respondent knows that not-p is true. Of course, UNDER TRCP
198.2(b) both denials admissions can be qualified if it is done in good faith
and  reasonable justification is set
forth in some detail; a five-word qualification is seldom sufficient.  A denial of p does not even entail as a matter of true, falsity, and evidence
that the party knows that p is true,
but it would be imprudent to admit a proposition unless one is certain that it
is true and strongly believes he has evidence to that effect. This combination is something like knowing. 

some cases denials are easy. “Is this your signature?” is often like that.  “Is this one sheet of paper the principal contract in
the deal, artless though it be?”  When
more complex topics are involved denials are not that easy.  This would be true of diseases, for example,
and psychological states.

The asymmetry between A and D just discussed can be softened when there is a middle ground between A and D. There is such a thing, and that is not able to answer “N”). Frankly it is not clear exactly what this is, or where N ends and D begins. One can lose a lot of sleep about this. 

How’z about this as a different approach: if the responding
party does not “know and had tried to figure the thing out,” it may be OK as a
qualification if there is are justification for not knowing and for the failure to figure out. My phrase, “figure things out,” is
ambiguous. It might mean “The client has tried to find the relevant information
but can’t seem to do so.” The alternative is “This request is incoherent so
that a rational person cannot give a univocal response.” This idea works best
when the Request uses ambiguous terms or is a complex sentence, though
slightly more difficult to deal with when the Request consists of one compound
sentence, where neither part depends on the other.

It must be stated that at
least sometimes “The client can’t figure it out,” is hard to distinguish from an
objection to the request.  What difference that might
make is unclear.  It seems obvious that a
sound objection can count as a reasonable foundation for why a party cannot
admit a fact. Ds (and Ns) require some reasonable inquiry or investigation. What
counts as one of those depends on the context. 
In some cases, it requires reviewing.

There is a paradox built into Texas Request for Admission rules.  On the one hand, the rule says, roughly “You can ask about anything relevant.”  The courts, however, has said that RqfAs are for that which is uncontroversial and that they are built to simplify and speed trials up. Many realize that these are are not actually built into the rules, though they are found in many reported decision. 

is confusion as to the true meaning of “Admit” and “Deny.” Proposition
p is specified to be conclusively
established for the purpose of a given lawsuit if proposition
p is admitted by a party. Thus, the true
meaning of “Admit” is that the proposition that has been “Admitted” has been
A “Deny” is not necessarily the claim that
is true. A denial is simply the assertion that the responding party is not
willing to stipulate
p  and
demands that evidence in support of p  be provided in the trial. It is a refusal to stipulate.
 (Interestingly, a proposition that is admitted
is not necessarily true. It is even possible that the respondent does not
believe that it is true. It is simply that the parties have agreed that
p should be taken to be true for the
purposes of the pending litigation.)

So there are two theories about Requests for Admission and Responses thereto.  The first is the standard theory.  The trouble with that theory is that the terms “admit” and “deny” are being used wrongly, and this fact creates unending doubt as to what to do and what to expect.  The second is the non-standard theory. In substance it is better than the standard theory. Its problem is that it departs substantially from the usual meaning of the terms “admit” and “deny.” Because of the size of this departure, a responding part using it is dangerous. 

There is, of course a third route, and that is to object to everything which is not “black-and-white” true or false.  This is really “can’t figure it out” in disguise. 

And/Or object to every proposition with respect to which the responding party is not dead certain. Watch out for this one though. . . .

*Michael Sean Quinn, Ph.D, J.D., Etc.
Law Office of Michael Sean Quinn
1300 West Lynn #208
Austin, Texas 78703
(o)(c) 512-656-0503