I would certainly hope that they clarify that a lot better than currently presented. The first paragraph regarding Enemy Units and Retreats states that when a unit makes a normal move, no part of the move can be within 3" of an enemy model. The paragraph immediately after runs counter to this ruling by stating that a unit starting a normal move within 3" must either choose to remain stationary or retreat. Call me a rules lawyer here, but the prerequisites for the latter statement would not be feasibly possible without a separate ruling that circumvents the first one (either by superceding it or by defining it as a separate move type altogether) instead of the conflicting catch-all we currently have.
Which is why I posted the FAQ earlier that elaborates on what specifically a "normal move" entails and the rules that govern it. That, in conjunction with what is laid out in the core rules, actually makes it a relatively well defined part of the AoS rules (if you can call any part of GW's rules well defined...) and quite intentionally includes a "retreat" as falling under the umbrella of "normal moves." The problem is that they use 'normal move' as a referential term and not a specific term. For example "make a normal move" isn't one specific thing you're doing. It's a range of things you can do (retreat, run, move) that all fall under "how does this model move in the movement phase" sort of broad umbrella definition.
I think you have to always assume path of least resistance. If blood knights could retreat and charge, why not simply state they can retreat and charge? I don't have a great answer for this.
The other question is, if this counts as a retreat, why not call it a retreat? This question can have several answers. A) The terms are interchangeable as defined by the Core Rules and subsequent FAQ. or/and B) There is further developments in 3.0 that will make this specific language clearer.
Since we already know charge reactions will be a thing and "retreat" was a charge reaction from fantasy, its not outlandish to assume that they are isolating that language of "retreat" to that charge reaction in 3.0.
That interpretation also avoids the strange situation that comes from trying to separate a "retreat move" from a "normal move." If those are separate actions, it creates A LOT of interesting scenarios for the several rules that allow "normal moves." For example, it means the hero phase move from FEC's Royal Mordants battalion can no longer be used to retreat... which seems to directly contradict what the FAQ outlines.
Simply put, the type of interpretation your friend is implying creates a lot of other problems if you extrapolate that same logic to the rest of the game.
