Pilot DAR
First off I think it is important to point out that while my principal flying job is flying an airtanker in aerial forest fire suppression operations I also hold a current Transport Canada Class 1 aeroplane flight Instructor rating and a Class 1 aerobatic instructor rating, so I can situate the theory into the practice when it comes to discussing instruction practices
I am also discouraged by the "dumbing down" of flight instructor ratings. This is IMO particularly egregious with multi engine training which is unfortunately taught by instructors which have little or no actual twin flying experience.
I think all flying instruction should be conducted with a consideration of risk and rewards vs the training objectives. An example of this would e your spin example. Obviously you can not demonstrate a spin without actually stalling the aircraft with enough yaw to allow the aircraft to depart and enter auto rotation. In the case of a VMC demo however the effect of VMC and the procedure to recover from a VMC induced loss of directional control can IMO be just as convincing be demonstrated by limiting rudder travel such that
the MC induced ya occurs at a higher safer speed. Therefore my personal risk reward calculus is that there is no benefit to talking the speed down to actual
VMC.
In any case the problem with VMC is not recovering from a preplanned and briefed VMC demo at altitude, a situation that is unlikely to actually occur in the real world, it is to deal with the low altitude and low speed regime in the case of an EFATO. There is a horrible record of low altitude fatal loss of control accidents after engine failures in light twins.
The current multi engine training syllabus is IMO not doing a very good job of preparing new twin pilots because it is simply too dangerous to do actual low altitude simulated engine failures. The good news is the emerging new generation of flight sims like the Redbird and the alsim now have realistic enough visual systems and flight modelling to allow students to practice the worst case scenarios. I strongly encourage all the light twin student pilots I know to invest in a few hours of training in one of these sims even though the tie can not be counted. It is also vital IMO that anyone flying a light twin take annual recurrent training. If they don't then I think the second engine will only be good for taking them to the scene of the accident......