There are two senses of the phrase "in order that" that we should consider. One is the sense that is used when we are talking of human purposes: "the building was wired with explosives in order that it should collapse in a controlled manner". The other is the sense that is used when we are talking about "divine" purpose: "I was afflicted with this hideous skin disease in order that I might learn the virtue of humility".
Both senses have it that a series of events was "ordained"; they differ as to the origin of the ordinance, and the type of agency involved in bringing it about. That which is divinely ordained might be brought about by "intervention", the exercise of supernatural but causally specific agency within history, or it may be simply be a matter of "the universe...unfolding as it should", fulfilling its predestiny. That which is humanly ordained may similarly be the result of discrete, deliberate actions, or "systemically" produced by particular forms of social organisation.
Now, it is not divinely ordained that replicators should replicate (a replicator is an entity that ordains, at a minimum, its own replication), or that species should evolve with particular characteristics: evolutionary theory's sense of what is "ordained" includes neither divine (trans- or extra-historical) nor human (individual or social-systemic) ordinance, but rather extremely local and short-term goal-execution that is a) massively parallel, and b) massively iterated (both for really very massive values of "massive"). However, it doesn't work without a split/correlation between "representation" and "instantiation", and a mechanism whereby the one "steers" the other. Claiming that it was without "purpose" in this sense would be like claiming that it was acausal: you'd be left with "blind chance" without any way of gaining traction in the material organisation of species. Evolution is exceptionally parsimonious with purpose: it involves the absolute minimum necessary to get things done; and it does so without there needing to be any larger purpose governing this distribution of low(est)-level purposiveness.
The question about Dawkins I think is whether he sometimes speaks as if there were some "larger" purpose in evolution than the most minimal purposiveness demonstrated by his "blind watchmaker". I'm not convinced that he does; but it's arguable at least.