I have been a student of incomplete things for most of my professional life, which is to say I have been a student of almost everything. A card trick that fails in the middle reveals more about its own mechanics than one that succeeds without comment. A sentence that stops before its period is often more illuminating than one that lands. The incomplete object shows you the skeleton. What bothers me about the Colorado fields is that the skeleton is consistent. Whatever is making these marks is not failing randomly. It is failing systematically, and in the same direction.
I first noticed the marks in the late 1880s, working a circuit through the southern front range. A rancher near Pueblo asked me to look at a section of his wheat field where the grain had gone down in a pattern he could not explain. I looked at it for twenty minutes. It was an arc — a partial ring, maybe a third of a full circle — with a spur running east from the open end. The grain was bent, not broken. Whatever had pressed it down had done so evenly, which takes either great patience or something other than patience. I drew it in my notebook and moved on. I did not have a theory.
In the years that followed I encountered four or five variations on that first mark. None of them completed the geometry they appeared to be attempting. A double ring near Alamosa. Three arcs in native grass near the old Bent’s Fort route, from a survey log I came across in a Bent County records office — those marks dated to 1868, which meant they preceded my Pueblo observation by two decades. An incomplete spiral near Del Norte that a rancher had preserved like a specimen, refusing to harvest around it, as if he understood that he was looking at something worth keeping even if he could not say what it was. He was right. He could not say what it was, and neither could I, but he was right.
I filed these observations and did not connect them to anything until I spent a season in England in the early years of this century.
England has its own fields, which is not surprising, and its fields have been doing something that its residents found alarming enough to write about extensively, which is. The formations in Wiltshire — and I walked through several of them on foot, which is the only way to understand their scale — are large, elaborate, and complete. Not complete in the sense that a sentence is complete when it says a plain thing plainly. Complete in the sense that a fugue is complete: internally organized, formally resolved, built on a structure that you can feel before you can name it. I walked the perimeter of one formation for forty minutes and arrived back where I started with the genuine sensation that I had not been moving in an arbitrary direction.
I recognized them. I recognized them the way you recognize a tune you have heard played badly for years when you finally hear a virtuoso play it correctly. The forms in Wiltshire are larger and more resolved than anything I had seen in Colorado, but the underlying geometry was the same geometry. The arc that stopped east of Pueblo was the same arc that completed itself in the chalk fields of Wiltshire. I had been looking at drafts. England was showing me the finished piece.
Now, I want to be precise about what I am not saying. I am not saying that England originated this pattern and Colorado copied it imperfectly. The Colorado marks are demonstrably older. The Bent’s Fort arcs predate the earliest English formations I was able to find records of by at least a decade. What I am saying is more interesting and stranger: the pattern appeared first in Colorado in an incomplete state, and later in England in a complete state, and this sequence does not suggest imitation. It suggests development.
I have watched many performers in my life, and I have watched many of them work on a trick for years before it was ready for an audience. The working version of a trick contains everything that the finished version contains — the same logic, the same mechanics, the same fundamental geometry — but it fails at the moment of resolution. The Colorado fields are the working version. Wiltshire is the performance.
Whatever is doing this, it learned.
I do not know what that means for us. I know what it means for the archive: the Colorado fragment log must be treated as primary evidence, not as a collection of agricultural curiosities. The Wiltshire formations are the confirmation. Colorado is the original text. And the fact that the original text is broken — that every Colorado formation stops before completing its circuit — is not a flaw in the record. It is the record. Something began here and did not know how to finish. Later, it did.
I leave the question of whether that is comforting to the reader’s judgment.
For my part, I note only that the open end of every Colorado arc I have examined faces east, and that the Wiltshire formations, when oriented on a map, tend toward the west. I have drawn this implication in my notebook and declined to put it into words. Some geometries are better left as diagrams.