backside attack
If you wanted to push someone out of a doorway, you wouldn't shove from the same side they're standing on — you'd come at them from behind. The SN2 nucleophile does exactly this: it attacks the carbon from the side directly opposite the leaving group, 180 degrees away. This geometry is called backside attack, and it is the heart of why SN2 reactions behave the way they do.
Why the back, and not the front? Two reasons. First, the leaving group's electrons and the lone pairs sit on the front side, electrically repelling the incoming electron-rich nucleophile. Second, and more deeply, the carbon's empty antibonding orbital (the sigma-star of the C–X bond, the orbital the new electrons must flow into) has its biggest lobe sticking out the back, opposite the leaving group. The nucleophile's electrons slide most smoothly into that backside lobe. So the new bond forms on the far side while the old bond breaks on the near side, the two motions perfectly coordinated in one step.
Backside attack has a direct, observable fingerprint: it turns the carbon inside out. The three groups that stay behind get pushed through to the opposite side, flipping the configuration like an umbrella in a storm (the Walden inversion). It also explains SN2's steric sensitivity — if the back of the carbon is shielded by bulky groups (as in a tertiary halide), the nucleophile can't get in, and the reaction stalls. Geometry, not just energy, dictates the outcome.
In a cyclic substrate, the nucleophile must reach the back of the carbon, so a leaving group locked in an inaccessible position simply won't undergo SN2 — geometry vetoes the reaction.
Access to the back of the carbon is non-negotiable for SN2.
Backside attack is unique to SN2. In SN1 the leaving group departs first to form a flat carbocation, which a nucleophile can then hit from either face.