That’s what the Kansas Supreme Court just held, interpreting Kansas Stats. § 21–3211. The statute reads,
(a) A person is justified in the use of force against another when and to the extent it appears to such person and such person reasonably believes that such force is necessary to defend such person or a third person against such other’s imminent use of unlawful force.
(b) A person is justified in the use of deadly force under circumstances described in subsection (a) if such person reasonably believes deadly force is necessary to prevent imminent death or great bodily harm to such person or a third person.
(c) Nothing in this section shall require a person to retreat if such person is using force to protect such person or a third person.
And the court concluded that while this allows self-defense that involves an actual attack on the attacker — for instance, hitting, shooting, or stabbing the attacker — it does not allow self-defense that merely involves a threat of violence against the attacker.
I think the dissent is right to say that “force” can reasonably be read as including “constructive force” such as threats, especially in light of the substantial American legal tradition of reading force this broadly (and despite the fact that other Kansas statutes generally do say “force or threat” or some such). And this is especially so because, as the dissent points out, the result is absurd: Restraint in the use of defensive violence is rewarded by criminal punishment. I believe courts should generally read statutes as written, but the should also read their terms against the backdrop of the legal rules that help define these terms, and principles such as the rule of lenity, and the presumption against readings that produce absurd results. ...
You can read this short, boneheaded decision here. The bottom line: in Kansas, you can use actual physical force to defend yourself (and be entitled to a jury instruction on self-defense), but not the threat of physical force (which, in the case of guns, has the statistically greatest deterrent effect). So the net effect of this decision is to discourage deterrence, and encourage actual physical violence. Dumb. Indeed, the dissenting opinion points out the absurdity of the majority's position.
As Professor Volokh suggests, the Kansas legislature needs to fix the statute to specifically allow the threatened use of force in a self-defense situation.
Indeed, I believe other states such as (I think) Montana and Arizona have recently changed their laws to specifically and explicitly allow so-called "defensive displays" of force in self defense.