I throw this out as a hypothesis, in case Stranger or someone can confirm/deny it.
A larger part of war than destruction, quite often, is the matter of logistics. For example, it is of no value to have the best gun in the world, capable of blasting a hole through any surface, if you can only make one a year and you have nothing that can move it around.
In the case of a nuke, I can envision three logistical “quirks” that would lead to the development of a variable yield weapon.
First, would be the question of what is the ratio of sizes to build? Do you want to build 10% small size nukes, 50% mid-sized nukes, and 40% large sized nukes or something completely different? Nukes are such a strange weapon that it might not be clear what the demand would be for usage, should the call ever come to start deploying these things for actual us.
Second, would be the matter that these are hard to build. Sufficiently hard that it may be easier to set up one production pipeline for a slightly more advanced weapon than trying to set up three production pipelines for multiple weapon sizes.
Third, would be the matter that there might be very little physical size difference between a “small” nuke and a “large” nuke. With C4, if you want a bigger boom, you add more C4. With a nuke, it might be more something like you adjust the timing on the explosives, but fundamentally you still end up with all the same parts in all the same sizes. And with that the case, where there’s no real physical need to construct the whole device differently for different yields, and it can be tuned after the fact, then why not just do that?
”An Army, like a serpent, goes upon its belly.” — Thomas Carlyle
While it is true that there is little difference in size and the espensive Primary material between at tens-of-kilotons range weapon and one in the hundreds or thousands of kilotons (since most of the greater yield comes from energy released or focus by the Secondary) the complexity of a multistage weapon and especially a variable yield weapon is not trival because you want it to provide a reliable yield at all selectable levels. I think all variable yield weapons (that I’m aware of) are based on modifications to an existing fixed yield design with features to allow or retard peak yield levels. This does allow for a more flexible response in battlefield conditions, but the United States has never given control of nuclear weapons to battlefield commanders; except for SDAMs and MDAMs, which were never operationally deployed, release of all nuclear weapons would still have to be authorized by a direct command from POTUS (except for contingeny plans in the case of a disarming first strike about which there is sparse publically available information). Variable yield just makes weapons like the B-61 and B-83 more flexible in terms of use, but again, the difference in even the most powerful conventional weapons compared to the least powerful nuclear weapons makes any notion of “proportionality” kind of risible.
Also, I’ll say that I just wish this was a purely hypothetical exercise of perverse Cold War nostalgia, but the W. Bush administration brought nuclear weapons back to the forefront with their abortive attempt to propose nuclear “bunker busters”, and given that Trump knows exactly fuck all about the nuclear triad (when you get schooled by Ben Carson on anything, you are really lacking in awareness) and yet is dumping treaties and levelling barely veiled threats toward countries like Iran and North Korea, discussion of the effects and potential failures of non-deterrent uses of nuclear weapons becomes extremely disconcerting to say the least. As if I didn’t have enough problems sleeping already.
All this is an extremely learned discussion in which many relevant facts are offered for consideration. It’s also astonishingly naive.
The real point of variable yield nuclear weapons during their heyday is for military contractors to cash big checks. The goal of any contractor - in the United States, anyway, though I’d bet on elsewhere - is to roll out something new, shiny and expensive just as their prior contract runs out. Got Hiroshima bombs? Oh, that’s fine FINE, sir. But someone else MIGHT be developing a FUSION bomb. Better pay us to get that puppy up and running. Hey, that might be too inflexible. How about something where you can set the amount of boom ahead of time. Blah blah blah ad nauseum.
While the statement that contractors promote opportunities for profit is not entirely untrue the general claim is also overly reductionist. The original demand to expand the nuclear arsenal and develop multistage thermonuclear weapons wasn’t pressure from the defense industry (which had essentially no experience with nuclear weapons because they had been developed by the US Army Corps of Engineers and later by the US Atomic Energy Commission, with the design/test labs being run by the University of California and the Oak Ridge and Hanford production facilities being run by University of Chicago. It wasn’t until after the war that the sites started to be managed by outside contractors, and design authority for new weapons has always been maintained by the LLNL and LANL rather than by production contractors, unlike other weapon systems like combat aircraft and naval vessels.
The design for variable yield nuclear weapons came from a perceived need for a more flexible response (and also justification for maintaining strategic bomber fleets at a time when reliable and capable ICBMs were being deployed, and SLBMs were showing equally flexible and capable responsiveness), which started at a time that nuclear war was still thought (by some) to be “winnable”, and frankly it is kind of obvious because making a variable yield weapon largely consists of adding or removing neutron enhancers/reflectors, so it isn’t as if they comprise an entirely new class of nuclear weapon design. In fact, most variable yield weapons are modifications of pre-existing weapons already in inventory. The development of the thermonuclear weapon by the Soviet Union—which was accomplished almost a decade sooner than most informed estimates predicted—is what drove the expansion of classes and modifcations of thermonuclear and boosted fission devices as it became apparent that the United States could not expect to win an all-out war by volume of the arsenal alone, especially when the Soviet Union also displayed prowess in the area of ballistic missiles (albeit with poor reliability).
Contractors have made a lot of money and cut corners in subsequent decades, largely in enriched uranium and plutonium manufacture, but it has always been perceived strategic need which drove the demand for new weapon designs and new delivery systems. That isn’t to say that contractors didn’t gig decision-makers in the government to push for new generations of missile systems and weapons (there was arguably little need for the Peacekeeper Rail Garrison system and even less for the Small (mobile) ICBM) but most of the weapon system advances made sense in the context of the day, e.g. the evolution of the LGM-30A/B/F/G ‘Minuteman’ and the Polaris/Posiden/Trident Fleet Ballistic Missile systems in terms of increased capability and accuracy.
There is no argument that we’d have been better off spending that money on developments that improved upon the lives and health of American citizens and stability abroad rather than building and testing nuclear weapons which exposed millions of people to small levels of radioactive fallout and produced a brinkmanship contest which brought the United States and Soviet Union to the cusp of war on multiple occasions, but a combination of distrust, misapprehension, and ideological blindness led to a contest neither nation felt they could back away from, and we can just be happy that incidents like the Cuban Missile Crisis and the Petrov Incident were averted by the calm thinking of a few key individuals and no small amount of luck. And frankly, the government is pretty parsimonious about spending money on socioeconomic and non-military technology improvements compared to the massive amounts spent on defense, often with unclear or negligible benefits, even though we have gained commerical technologies arising from military research (robust and reliable microelectronics, space launch capability, the Global Positioning System). But reducing the origin of variable yield weapons and many other military technologies strictly to profiteering is itself naive and overly simplistic.
The U.S. did deploy nuclear warheads that could withstand the shock of being fired from artillery. (Wikipedia lists the U.S. W9, W19/23, W33, and W48) Caveat: I think only “Atomic Annie”, firing a W9, was the only test that fired a nuclear shell from a gun tube.
Naval guns fired shells that had a delayed fuze feature that gave the shell time to fully penetrate (warship) armor before exploding. This has been a common feature of AP shells since the 19th century.
Do you feel that the nuclear warheads could NOT be adapted into some kind of ground penetrating role? Or are you merely stating that we don’t have any at the moment?