EARLY CLASSIFICATION SYSTEMS FOR U F O PHENOMENA

To my knowledge, the first classification system for UFO phenomena grew out of my work in 1961 and 1962 with Aime Michel and with Dr. Pierre Guerin in Paris. I proposed to separate the massive collections of French sightings into four major categories. Briefly summarized, they encompassed the following:

Type I, which was divided into three subgroups, gathered all the landing reports that Allen Hynek would later call close encounters. It may be hard for the modern reader to realize that until the late Sixties American ufology did not admit that such reports even existed. Among the major civilian organizations, only APRo, headed up by Jim and Coral Lorenzen, recognized their significance. The Air Force's Project Blue Book automatically sent landin~ reports into the psychological category.

Type II reports were observations of the cloud cigars so prevalent in France in 1954, and whose role in the development of major waves had been discovered by Aime Michel. It is a source of puzzlement to me that these reports have practically disappeared from the scene. I am unable to speculate about a reason for this change. But the fact is that this category is no longer justified.

Type III were those sightings of objects that had a discontinuity in their trajectory; either they stopped and hovered, or they came toward the earth with a falling leaf motion before resuming their flight, or they performed some maneuver that identified a specific point in space and time. This was important to me since I was compiling a case catalogue in which I wanted to include precise longitudes and latitudes.

Type IV, by contrast, gathered all cases of oobjects in uninterupted flight.

This classification served its purpose for many years. It enabled us to discover specific patterns for various types of UFo behavior, notably the law of the times for the landing reports.

In 1972 Allen Hynek built upon this fin in The UFO Experience to divide all reports into two groups: the short-range sightings that corresponded to my Type I and which he called close encounters, a term later immortalized by Steven Spielberg; and the objects observed "at some distance," which he further divided into three categories:

1. Nocturnal lights for objects seen at night (NL)
2. Daylight disks for those seen in daytime (DD)
3. Radar-visual for those detected on radar (RV)

Hynek was well aware that this division was arbitrary, and he remarked himself that the categories "may not be mutually exclusive."

The major contribution in Hynek's classification was the clarity with which landings were now defined:

• CE1 was the class of objects seen on the ~round or at short distance to the observer.

• CE2 was the class of close encounters in which physical effects or traces were present.

• CE3 was the class ol close encounters that involve "entities" or "occupants," a category from which he excluded repeated contact.

These categories have withstood the test of time and continue to be useful today. In recent years a fourth category has been added. Called CE4, it encompasses the abduction reports in which the witness has not only seen the occupants but claims to have extensively interacted with them inside their craft.

Serious problems arise, however, when one tries to use Hynek's categories of nocturnal lights, daylight disks, and radar-visual cases, especially when the screening process works from a computerized database, as it will need to do in any sophisticated effort to attack the UFo problem in the future. Not only do those categories overlap, as Hynek had noted, but many reports cannot be placed within any of the classes.

Examples of the former problem are provided by the numerous cases of disks seen in daylight and tracked on radar at the same time. If they are placed into the RV category, the statistics on DD cases are amputated by one potentially important data point, and vice versa.

An example of the latter deficiency is Whitley Strieber's observation of a dark disk flying between him and the field of the stars at night: this was neither a daylight disk nor a nocturnal light.

Other problems arise with daylight objects that are not disks, like the flying hexagon seen by Mr. Dillon in 1928, and with the cases that are observed at dusk or at dawn, and which could conceivably belong to several categories.

The work we have done with computer-based expert systems in the last three years has suggested new, more practical, solutions to these remaining problems, although it must be recognized that no classification system is perfect.