I tend to agree with Michael’s thinking, it is certainly not a true arch, and I don’t doubt his dry blocks experiment. Maybe the word he and I want is ‘a long, high, corbel, and he said as much with his canti. on canti. thought. Certainly the corbel has some horiz. tension component at its top and some horiz. compression (thrust) component at its base. I wouldn’t completely ignore some thrust, since I suspect for a few courses at the top there will be a compressive reaction in those brick courses, at the meeting of tips of the facing corbels, but not to the extent of a keystone in a real arch. I would like to see some continuous, in plane, vert. & horiz. veneer at my lintel bearings, to take whatever the reactions are.
I have seen pictures and field examples of what JAE talks about. Obviously, a real arch over the opening, in a brick veneer, will have more horiz. thrust reaction than the opening with a structural lintel under the brick. The lintel is their, at least, to carry the triangle of brick that Michael says falls away (an unreliable function, in tension, of bond btwn. brick and mortar), when the false work is removed. I agree with JAE re: friction at the steel lintel, but that’s called bond btwn. the bricks as we talk about the corbel.