To answer the comment about a "bow wave" effect shielding components from ice accretion.
There can be an element of this present, and it would be particularly noticeable for fuselage mounted antennae, for example. But it's simply not a large enough effect to shield the tailplane. (Consider the fact that engine anti-ice is routinely fitted to aircraft with aft-mounted engines).
Additionally, the inflight evidence is clear - ice does accrete on virtually any surface in the airflow. The key for the aircraft with no tail de/anti-ice capability is, as stated before, that they are DESIGNED with that consideration in mind. It's a tradeoff between the complexity (and weight) of another de/anti-ice system, versus the cost and weight of an otherwise oversized tail surface. Generally the latter tends to be the better compromise.
I can't find any of our icing pics in the public domain right now, but NASA has some impressive pictures in their icing training.
The following quote is from intro #2 of their
In-Flight Icing Training
Originally Posted by DC-10 pilot
As we arrived at the gate, I noticed that all the mechanics were standing off to the side, pointing at something in the back of the airplane. After the people got off, we went to the rear of the airplane and noticed there was about eleven inches of ice on the horizontal stabilizer.
I think that's fairly conclusive in showing that tailplanes of large swept wing types can, and do, accumulate ice.