They were Ukraine's in the sense that they had physical possession and unlimited time to circumvent the locks ("permissive action link", PAL). It's true the PALs may have been very non-trivial to circumvent (e.g., some PALs require a code to decrypt firing parameters), but I haven't seen anyone argue that this wasn't still vastly easier than developing weapons from scratch. Ukraine had substantial nuclear weapons expertise; it was not just a physical hosting site of the weapons.
It is hard to guess because politics is complicated and it is a hypothetical, but I'd be surprised if they really had unlimited time to circumvent the locks. If they hadn't given them up peacefully, for all we know somebody might have decided to take them by force.
Can you give some examples? I brought up Chechnya because it declared independence literally at the same time as Ukraine, and Russia waited for 3 years before finally sending the troops in (and that turned out to be a huge mess!).
Hitler mainly. Napolean? those are the classic examples. Both thought Russia was an impotent land of peasants. It's not like Russia has been invaded that many times.
If soviet russia's PAL locks were built anything like US's security measures where it's actually subtle timing information they would have to rip out and completely rebuild the explosive shell that trigger the nuclear reaction, at which point you essentially are creating new nukes from scratch.
1. It's not obvious to me you can't reverse-engineer the necessary timing information, especially when you have a bunch of nuclear weapons experts and a huge stockpile allowing you to take apart a few and use that knowledge on the rest. To my knowledge, the PAL system was intended to prevent weapons from being detonated by rouge officers or thieves; it wasn't designed to thwart states.
2. Even if all the conventional explosions are worthless, you still have all the nuclear material and other extremely sophisticated equipment involved with boosting, etc. It's not like creating new nukes from scratch.
I think for this conversation to continue productively, we need experts to weigh in on these issues.
They're not movie bombs where doing something wrong sets them off. PALs were very much the fail safe kind of device because if messing with them could set off the warhead then you've just discovered the way to detonate the warhead which a PAL was designed to prevent from happening. Nukes are also relatively safe in that it requires a very carefully timed sequence of explosion to properly implode the core so an accidental triggering is just a dirty bomb instead of a nuclear detonation.
It depends on what level the detonation was triggered at. An accidental detonation that actually triggers the primary device detonation mechanism could result in the bomb actually going off.
If an accidental detonation means just triggering some of the explosives, then yes, you'd get a dirty bomb. However, it's actually more likely that the bomb would actually detonate than the explosives being partially set off... they're designed specifically to be hard to detonate unless the actual detonator mechanism is used.
Honestly, the biggest risk in failure is that the missile gets launched, but the payload is a dud. So now, your country gets the punishment for launching a nuke, but none of the benefits.
Mostly if you're not smart enough to work remotely in which case you're probably not the person to be trying to bypass a PAL in the first place. And again they're not designed to trigger the bomb incorrectly if messed with because the whole point is to make them unusable unless you have the PAL code.
One might easily imagine an energetic disabling event that doesn't result in nuclear yield. Just fire one detonator.
That said, Command Disable mechanisms on modern weapons don't result in a loud bang. However, there's no guarantee those mechanisms are all that are used in mechanisms which may deter physical penetration of vital areas of the weapon.
It's easy to imagine but also it makes these immensely more dangerous to house and transport. Why fail dangerous when scrambling the timing information is as effective at preventing a nuke from being used?
Or the nuke does an intentional fizzle (sets off the implosion in an intentionally asymmetric manner resulting in plutonium being spread all over, but no nuclear criticality) which seems like the most serious-but-still-plausible "fuck you for trying this" mechanism I can think of.
How much of nukes is even hardware vs software? I'd naively think that you'd probably just rip out the core of the device and wire up a new thing that can zap it to go boom.
Nuclear bombs are strongly disinclined to "go boom" unlike say TNT. Turning the energy from splitting atoms into a large explosion, which is what you want from a weapon, will require precise timing. If you get it wrong either nothing happens, or you maybe create a small detonation, and cover a modestly sized area with dangerous debris from the failed attempt. Just throw a box of grenades into a waste water treatment plant or something instead for a fraction of the cost.
For reference, there have been a number (dozens, at least) of nuclear devices dropped accidentally all over the world (some of which landed in the USA), but none have ever accidentally detonated.
This was a plot device in the popular Fallout games. Where a city had built up around a nuclear bomb which failed to detonate during a nuclear war.
The timing of the explosives around the core is extremely important to actually getting a nuclear detonation instead of just a dirty bomb so the precise timing and triggering of the detonators is a very critical part of the bomb. They're not just simple fuses you can light and run away.
Then use rad-hard arduino/esp-32/cortex-m0/RittzkenfaiiV/whatever equivalents running NTP / Precision Time Protocol(PTP) in a cluster, with the elements embedded in the individual explosives. Maybe read about https://duckduckgo.com/?q=+fourth+generation+nuclear+weapons and how that could be applied to the material at hand, Ka-Boom!
You can, but it's not as simple as it sounds. Since the timing of the explosives is critical to achieving a nuclear explosion, you can't really do the timing all in software. You need a specialized network of switches that splits the detonation signal several times so that all the explosives go off at exactly the same time. Even the wires have to be cut to the same length with a tiny margin of error.
You can replace the timing and electronics mechanism prior to that network, but probably any sort of tamper resistant mechanism for a weapon will remove part of that network of switches if removed from the device. The rest of the controls are just too easy to replace to be effective at keeping someone from a roll your own type solution.
I think about nuclear physics and high performance aerospace as the distinction between hardware and software shrinking to zero.
You're effecting physical operations (movement, explosions) on such tight timescales that the software becomes part of the hardware. You can't just run the code on a different setup: the code is defined by the hardware it's running on, because it's orchestrating the physical properties of the hardware it's running on.
As an analogy, think about programming early video game systems or computers, where a single clock cycle was critical. Is the software just software? Or is it intertwined with the hardware it's running on? (See: emulators having to mimic actual hardware performance)
