In 1993, everyone knew that "text adventures" were a dead genre, but, as has been lovingly documented by Jimmy Maher and others, a few people here and there labored in obscurity to keep the nascent art form of interactive fiction alive. Art LaFrana was one of these, someone who eschewed the semi-commercial offerings of the time such as TADS and crafted his own custom system for MS-DOS.
The Abbey, LaFrana's second work (which is not to be confused with the 2008 title of the same name by Steve Blanding), received more attention than might be expected for a shareware entry with an extremely primitive parser. A review in SPAG Magazine by Cedric Knight suggests that it was considered worth paying for even 8 years after its release. I don't think I would have agreed at the time, but in an era when so few new games that even looked like a text adventure were being produced, it may have seemed a more attractive option than it does today.
To say that the game has a parser is being somewhat generous; my own interaction suggests that it is a simple keyword-matching system that doesn't really try to parse the player's command in any meaningful way. For example, a command such as >PUSH WAGON EAST will be interpreted the same as >PUSH WAGON or even just >PUSH. This doesn't matter as much as it might because the ratio of interactable objects to rooms is alarmingly low by modern standards, and the majority of the player experience is simple room navigation.
Other reviewers stress the text's ability to evoke an atmosphere, and I have to admit that I also found this to be the case -- but I don't understand it. It's really quite puzzling; the text is "evocative" in a paradoxical way in that on first reading it seems flat and uninspired but after extended exposure a fairly vivid mental picture of the environs emerges in one's mind. I'm not sure how much I can credit the author for this; in some ways it almost seems akin to the phenomenon of hallucinations experienced during sensory deprivation. (Is this the part that I've never understood about the appeal of Scott Adams games?)
Although the game is listed under the "Historical" genre, it certainly doesn't seem to be a very accurate portrayal of its ostensible period. (For example, the language of NPCs and signage can be anachronistically modern.) That's not to mention that Atlantis seems to have been a real place in the game universe.
It does seem more intent on historical accuracy with respect to architecture and living conditions. A large number of basically empty rooms are used to describe the sights and smells of the abbey complex, and I assume that this aspect is modeled after some real historical place -- there just doesn't seem to be any other motivation to include non-functional locations such as a pigsty. I have no special historical expertise by which to judge its correctness here, but it does feel believable.
The puzzles were to me the bad kind of old school, being based largely on intuition (which might not match the author's) and/or requiring a long series of essentially unmotivated steps without intermediate feedback. The largest part of player effort by far will be spent on creating a map, which is an essential step because room descriptions sometimes omit exit listings. Since I was trying to play this on a schedule (for the People's Champion Tournament) and wasn't finding anything to savor, I resorted to a walkthrough after a few hours, and I don't regret doing so.
I was ready to like this one, but in the end I found it to be a poor substitute for a proper parser, player-friendly puzzle design and minimally-interesting story. I'm tempted to go with one star, but, as mentioned, in the end it did end up engaging my imagination, and that part was enjoyable. I can't say I'd recommend it other than as example for someone studying the earliest examples of the form. Its most lasting impact might, ironically, have come from Knight's review: Inform 6's Standard Library 6/11, released afterward, implements many of the variations of the verb "pry" that he complained were missing in this game.