Anomalies Catalogue

"The most exciting phrase to hear in science, the one that heralds new discoveries, is not 'Eureka!' but 'That's funny ...'." - Isaac Azimov

As we have said elsewhere, our real aim is to find out how the world works and what our place in it is. Anomalies are useful clues because they point to areas where current scientific models may be failing. There is at least enough credible evidence to suggest that the world may be more complex than it is generally regarded to be. We feel that the data should be borne in mind by those scientists developing world models, as it may provide clues to help them in their theory building.

The trouble is that orthodox scientists, even ones sympathetic to anomalies research, have their own fields to worry about, and no time to become expert in the anomalies literature. The evidence is vast and often inaccessible in paper journals and foreign languages. It is varied in quality and some is hotly debated. Since the best evidence rarely attracts criticism, it often slips from view.

Even amongst anomalists and psi researchers, people tend to specialise in a few fields and may know little about other anomalies. We feel that it is vital to look at all the evidence jointly to try to make sense of it. After all, any model of the world has to be able to explain the entire range of observed phenomena.

So, we need a way of collating the evidence, sifting out the best and making it easily accessible. This is not a new idea - various proposals and attempts have been made in the past, but new technology makes it much more practical than before.

The old way of tackling this problem was to publish case collections and retrospective reviews. Deciding which cases to include was a one off yes-no decision, and once made, that decision is fixed. The choices made often reflect the biases and models of the age, and quickly become out of date as new evidence emerges. Furthermore, the cases are grouped in some way, and this grouping cannot be rearranged because of the constraints of the paper medium.

Nowadays we have databases built for large quantities of data. They are like index cards that can be instantly rearranged at will. An internet search engine is a good example of a database containing a huge amount of information that can be searched quite flexibly. A database of cases will allow you to create a case collection on the fly, tailored to your specific question, sorted and grouped appropriately. You can also tie different types of information together temporarily, so a case could be presented together with the critical arguments for and against, related cases, correspondence or experiments.

The important thing is to be able to ask questions that the database designers never anticipated. Fortunately, modern developments in information technology are geared to being able to solve exactly that sort of problem, since many industries have the same requirement.

Having decided that we want an anomalies database, there are a number of components to the task of building one. We describe these in the next sections.

Classification

For a start, we need to define and clarify the terminology that we are using to categorise cases. There are a number of interesting correlations between phenomena that are usually considered to be quite different. Consider, for example, dowsing and mental mediumship. Superficially, they seem completely unrelated, but ultimately both methods provide information that cannot be accounted for by any normal means. Some dowsers have been shown to predict the depth, strength and quality of boreholes more accurately than geologists can in areas where geological indicators are absent (Betz, 1995). Some mental mediums can produce information about people, under double blind conditions, that is shown to be far more relevant to them than to a control group (Robertson and Roy, 2001).

Many dowsers think that they are responding subconsciously to minute variations in the earth's magnetic field (although there is evidence to cast doubt on this interpretation). Mental mediums say that they get their information from departed spirits of previously living humans. We feel that the best thing to do, for now, is to concentrate on what the phenomenon does (i.e. present anomalous information) rather than what it appears to be (which is subject to interpretation). In other words, we focus on the resulting evidence rather than the context within which it was derived.

This approach is essential for the anomalies catalogue if it is to be used as a research tool. We need to be able to present the core evidence without constraining future researchers with our own assumptions and interpretations.

With this in mind, we have developed a new classification system as part of the first phase of this project. This is described in Challenging the Paradigm Systematically: A New and Generic Approach to Classifying Anomalous Phenomena published in the Journal of the Society for Psychical Research (JSPR), April 2002 [see abstract as HTML or paper as PDF].

Rating Cases by Various Criteria

The trouble with just putting a load of cases into a database, is that you may get hundreds of cases presented in response to a query. How can you determine which is most likely to be relevant to you, without having to read them all? If every case had a quality score associated with it, you could ask the database to only show you cases with scores above a given threshold. Then different researchers could cut off the scores at different places.

A single score is too constraining, however, because different researchers will place value on different criteria for scoring. It would be much better to have many different scores relating to different things. This would allow researchers to vary the way the database sorts the cases it locates for them.

At a high level, the different aspects of a case that we could score fall into four main categories:

• Case credibility : how credible is the evidence for this individual case?
• Category credibility : how credible is this case's category, i.e. how credible are the best cases of this type?
• Category size : how many similar independent cases are there in the category?
• Case value : how telling is the case - does it have specific features that have particularly significant implications?

Depending on the research you are doing, you may, for example, choose to focus on cases with significant implications (while lowering the criteria for evidentiality), or you may choose to look at only the most well established cases.

The next step is to identify a complete list of criteria that may be relevant to anomalous cases. Obviously some criteria may be more relevant to some categories of cases than to others. It will therefore be desirable to determine a criterion weighting for each category.

This project is currently underway, and to kick off the project, C-FAR organised an SPR Study Day, Making the evidence count, in April 2002. The speakers were invited to help brainstorm some ideas, with extremely fruitful results. We are in the process of developing some of the ideas that emerged from the day, and will be publishing the results here as they coalesce.

Database Design

It would be premature to do much work in designing the database before making more progress with the rating system described above. However, we will soon be in a position to start tackling this.

Data Collection

Although data collection is daunting in scale, there are no technical or conceptual difficulties in doing this. We hope to forge alliances with the many individuals and organisations that have created the case collections that exist. We also plan web applications that will allow the broader internet community to get involved in our project and help us with this work.

Application Design

The last phase of the project will be to build the applications that allow users to query, filter and sort the data. These will be multi-tier, component-based and open, so that they can be extended as new search techniques emerge. Application development will proceed in parallel with the data collection and loading phase.