Well, certainly you want to avoid accidental criticality. Not a lot of people know this, but in addition to the consideration or critical mass (which everyone seems to have heard of), there is the other half of the equation, that being critical geometry. In fact, there is a whole subfield of the engineering discipline devoted to criticality safety.
Now, that said, certainly steps are taken to prevent credible accidents leading to this. That involves placement of materials in such a way that accidental rearrangement of them does not lead to a critical assembly. Often, this is simply a matter of separation. Not 20 feet from where I am sitting now there is a subcritical graphite assembly with no control rods. Subcriticality is maintained by the pitch of the fuel rods being greater than a specified distance. So, imagine a series of physical separations between groups of fissile materials being emplaced to effect a safe geometry.
Of course, an added level of criticality safety can be had by incorporating neutron-absorbing materials in the separations. This can be very simple stuff, like natural boron or boron-bearing materials (e.g., boral plate). The Yucca Mountain designs feature both.
But, imagine if these safeguards broke down, and, by some chance, moderating material was introduced in and around the stored fissile materials (which the Yucca Mountain studies show hasn't happened in those formations for hundreds of millions of years). In spent fuel, after a long decay time, assuming the 3% enrichment typical of commerical fuel, it is likely that the fissile form of interest will be 235U. Its half-life is on the order of 700 million years, so its going to be around longer than 239Pu with its relatively insignificant (in terms of geologic time) half-life of 24,000 years. So, what would happen? Well, you certainly would not have a nuclear explosion, the fuel density and of course any reasonable means of attaining symmetric compression of the mass is just not credible. But, you could end up with a situation similar to the fossil reactors discovered in Gabon. There are several at Oklo and another at Bangombe. These operated off and on for millions of years in earlier geologic epochs, owing to the higher natural abundance of 235U in those earlier ages. The uranium in those formations underwent fission and, horror of horrors, actually produced fission products in water-bearing strata. Now, did the continent of Africa become a barren, radioactive wasteland as a result of the release of these uncontained, unregulated fission products? Hardly. In fact, last I checked, the place was downright verdant. Why? Because these fission products, even when uncontained in rock formations with lots of water flowing through them, only migrated a few centimeters from the site of the fission process. These are elements that just don't get out much. They tend to stay put and undergo their slow decay process, giving off an alpha or beta particle now and then, minding their own business, bothering no one.
So, bottom line, the Yucca Repository designs certainly allow for the chance of something happeneing, but even if everything went to pot, Hillary! would not have to have anyone around to do something "for the children" about it.
I digress again. Could this expain the evolutionary jump genetically to modern man and the lack of the missing link?
Food for thought.