Evolving Adapatibility

Adaptability is a sophisticated form of intelligence. Ergo, rats are intelligent!

I've criticized  de-facto IDist Cornelius Hunter  on at least a couple of occasions in this blog (See here and here), although I have conceded that he brings up some robust challenges to conventional evolution.  Along the latter lines I found his latest post very interesting. Quoting the first part of his post: 

Organisms adapt to environmental challenges. In fact, many different organisms adapt in non-homologous ways to many different, unforeseen, environments. This contradicts evolution. For we are not talking about random changes occurring by chance, occasionally getting luck enough to confer an adaptation, and then propagating throughout the population. We’re not talking about an evolutionary process of random mutations and natural selection. That would take a long time. What we’re talking about are adaptations that specifically address environmental challenges, and occur in a good fraction of the population, over a few generations, or perhaps within a generation. Such directed adaptation occurs quickly.

That contradicts evolution because random mutations are not going to create such a complicated adaptation capability. Furthermore, they are not going to do this over and over, in so many different species, for so many different environments. And even if, by some miracle, this did occur, it would not be selected. That is because the adaptation capability is not for the current environment the organism faces, but for an unforeseen, hypothetical, future environment. The moment it arises, the adaptation capability is of no use, and would not be selected for.

Hunter goes on to claim that these rapid adaptations also appear to be passed on to future generations whilst they are needed.  Unless the observational claims about an apparently inherent rapid ability to adapt (and pass it on for the duration) are false then this poses problems for conventional evolution, a process whose "look-ahead" ability is nil. 

Measuring Nth order differentials across time in the parameters of real situations is always  tricky because one needs to take samples that are increasingly separated as the order of the differential increases.  That is, Nth order differentials compromise the paradigm of locality, because an Nth order differential can not be calculated from data taken at a single point.  This is related to the questions Hunter is raising about the evolution adaptability: Adaptability suggests we have an inkling of what's ahead; either that or the change has been seen before. If it's the latter then using very local trial and error random changes requires huge numbers of trials. So unless the spongeam is extremely constrained the evolution of adaptability is a pretty problem for random trial and error. 

I find the following remark by Hunter particularly interesting:

What we’re talking about are adaptations that specifically address environmental challenges, and occur in a good fraction of the population, over a few generations, or perhaps within a generation. Such directed adaptation occurs quickly.

On the face of it this seems to be saying that organisms have been observed to adapt but this adaptation is not sufficiently explained by real time trial and error selection. If organisms do have an ability to adapt very quickly in way that can not be justified by resort to random trial and error process, then what is the source of this ability? Is it some built in mechanism that has, in effect, been learnt from many trial and error events in the past? If the latter then it would mean that the current situation, or something like it, has seen before. The alternative, however, is extraordinary and involves immanent intelligence: Namely, that some look-ahead mechanism is in operation. Is this evidence of backloading?

But if adaptability is a high level form of intelligence then is evolution itself intelligence in operation?