The Harvard station is rather special among MBTA Red Line stations. It’s a double-decker station, and also much, much curvier than other stations. But most notable to me—and, as I’ve found out, not just me—is that the station manages to evade geographical sense. On the inbound platform, my navigational intuition tells me the train should be coming from the right, but instead the train comes from the left. There are also others among the geographically-inclined that agree that it really feels like the train should come from the right. And yet, this is not just a small discrepancy in intuition, where a direction is a mild angle off. This is a navigational intuition failure of 180 degrees.
The entrance I (and most of my friends) usually take to go down to the Harvard station platform starts right in the Harvard Square bend of Massachusetts Avenue (which the Red Line runs under), heading eastward, along the avenue. After going down an escalator or flight of stairs, the path turns sharply left, heading down another escalator or flight of stairs. Here, the path branches into two curved paths heading out: one to the left, heading to the 71 and 73 buses, and one to the right, arcing over to the platforms, along the way splitting into one path going to the outbound platform and one going to the inbound platform, which is under the outbound platform. So, like this:
Thus, one would expect, after mentally processing this path, that the train comes from the right. Yet the train actually comes from the left.
There aren’t floor plans of the station available online, at least as far as I can tell, so I returned to the station recently just to figure out what’s going on once and for all.
It turns out this is what actually happens.
That is, of the 180-degree discrepancy, only about 90 degrees are from miscalibration of angles and curves in the path: the first turn is sharper; the second turn is shallower. The rest of the discrepancy is due to where the platform actually is being only about 90 degrees offset from the mental model of where the platform is, specifically, before the bend in the route. And in fact, looking into the tunnel towards the right at the Harvard station platform, one can confirm that the path taken by the train turns left after entering.
Interestingly, once I entered the platform with my mind set on figuring out what’s really going on, the explanation unraveled itself without need of additional tools or a map. That one time, focused on the problem, was more useful to entangling this mystery than the entire 5 years prior during which the station just caught me off guard.