The counterarguments to the "total war" doctrine are nuclear deterrence and asymmetric warfare. Total war is the correct way to win a symmetric war against an opponent you can actually defeat. But nuclear deterrence creates big incentives to not ever bother. Everybody loses from unrestricted warfare between nuclear powers.
Asymmetric warfare is destined to outlast the will of the attacker to maintain totally war, and to continue long after the attacker has started to become war-weary. IMHO, winning an asymmetric war against a sufficiently motivated opponent is only possible by engaging in genocide and large scale population relocations.
Using nuclear deterrence as an example. Yes, it is effective at avoiding battle but if you get nuked you must effectively wipe the aggressor's country off the map so it can continue to deter future aggressors and MAD can continue to be relevant. You can't hold back because of the enormous civilian casualties. The only thing important is to make sure that will be the last nuclear war and that no one thinks about using nukes against you. This is entirely possible for example with US vs NK or maybe even China (or if they ever achieve effective icbm interception...anyone)
The scary part of nuclear deterrence is indeed whether it is able to continue to prevent nuclear warfare. As you imply, technical innovations can change the picture. And nuclear warfare with limited escalation remains a scary possibility. Especially between countries that don't have MAD capability. Let's be glad that so far cool heads prevailed.
Asymmetric warfare is destined to outlast the will of the attacker to maintain totally war, and to continue long after the attacker has started to become war-weary. IMHO, winning an asymmetric war against a sufficiently motivated opponent is only possible by engaging in genocide and large scale population relocations.