Philosophers like Carl Hempel and Ernest Nagel regard a theory as scientific if it is hypothetico-deductive. To them, scientific theories are sets of inter-related laws. We know that they are inter-related because a minimum number of axioms and hypotheses yield, in an inexorable deductive sequence, everything else known in the field the theory pertains to.

Explanation is about retrodiction - using the laws to show how things happened. Prediction is using the laws to show how things will happen. Understanding is explanation and prediction combined.

William Whewell augmented this somewhat simplistic point of view with his principle of "consilience of inductions". Often, he observed, inductive explanations of disparate phenomena are unexpectedly traced to one underlying cause. This is what scientific theorizing is about - finding the common source of the apparently separate.

This omnipotent view of the scientific endeavor competes with a more modest, semantic school of philosophy of science.

Many theories - especially ones with breadth, width, and profundity, such as Darwin's theory of evolution - are not deductively integrated and are very difficult to test (falsify) conclusively. Their predictions are either scant or ambiguous.

Scientific theories, goes the semantic view, are amalgams of models of reality. These are empirically meaningful only inasmuch as they are empirically (directly and therefore semantically) applicable to a limited area. A typical scientific theory is not constructed with explanatory and predictive aims in mind. Quite the opposite: the choice of models incorporated in it dictates its ultimate success in explaining the Universe and predicting the outcomes of experiments.

III. Parapsychology as anti-science

Science deals with generalizations (the generation of universal statements known as laws) based on singular existential statements (founded, in turn, on observations). Every scientific law is open to falsification: even one observation that contravenes it is sufficient to render it invalid (a process known in formal logic as modus tollens).

In contrast, Parapsychology deals exclusively with anomalous phenomena - observations that invalidate and falsify scientific laws. By definition these don't lend themselves to the process of generation of testable hypotheses. One cannot come up with a scientific theory of exceptions.

Parapsychological phenomena - once convincingly demonstrated in laboratory settings - can help to upset current scientific laws and theories. They cannot however yield either because they cannot be generalized and they do not need to be falsified (they are already falsified by the prevailing paradigms, laws, and theories of science). These shortcomings render deficient and superfluous the only construct that comes close to a Parapsychological hypothesis - the psi assumption.

Across the fence, pseudo-skeptics are trying to prove (to produce evidence) that psi phenomena do not exist. But, while it is trivial to demonstrate that some thing or event exists or existed - it is impossible to show that some thing or event does not exist or was never extant. The skeptics' anti-Parapsychology agenda is, therefore, fraught with many of the difficulties that bedevil the work of psychic researchers.

IV. The Problem of Human Subjects

Can Parapsychology generate a scientific theory (either prescriptive or descriptive)?

Let us examine closely the mental phenomena collectively known as ESP - extrasensory perception (telepathy, clairvoyance, precognition, retrocognition, remote viewing, psychometry, xenoglossy, mediumism, channeling, clairaudience, clairsentience, and possession).

The study of these alleged phenomena is not an exact "science", nor can it ever be. This is because the "raw material" (human beings and their behavior as individuals and en masse) is fuzzy. Such a discipline will never yield natural laws or universal constants (like in physics).

Experimentation in the field is constrained by legal and ethical rules. Human subjects tend to be opinionated, develop resistance, and become self-conscious when observed. Even ESP proponents admit that results depend on the subject's mental state and on the significance attributed by him to events and people he communicates with.

These core issues cannot be solved by designing less flawed, better controlled, and more rigorous experiments or by using more powerful statistical evaluation techniques.

To qualify as meaningful and instrumental, any Parapsychological explanation (or "theory") must be:

1.. All-inclusive (anamnetic) - It must encompass, integrate and incorporate all the facts known. 2.. Coherent - It must be chronological, structured and causal. 3.. Consistent - Self-consistent (its sub-units cannot contradict one another or go against the grain of the main explication) and consistent with the observed phenomena (both those related to the event or subject and those pertaining to the rest of the universe). 4.. Logically compatible - It must not violate the laws of logic both internally (the explanation must abide by some internally imposed logic) and externally (the Aristotelian logic which is applicable to the observable world). 5.. Insightful - It must inspire a sense of awe and astonishment which is the result of seeing something familiar in a new light or the result of seeing a pattern emerging out of a big body of data. The insights must constitute the inevitable conclusion of the logic, the language, and of the unfolding of the explanation. 6.. Aesthetic - The explanation must be both plausible and "right", beautiful, not cumbersome, not awkward, not discontinuous, smooth, parsimonious, simple, and so on. 7.. Parsimonious - The explanation must employ the minimum numbers of assumptions and entities in order to satisfy all the above conditions. 8.. Explanatory - The explanation must elucidate the behavior of other elements, including the subject's decisions and behavior and why events developed the way they did. 9.. Predictive (prognostic) - The explanation must possess the ability to predict future events, including the future behavior of the subject.

10.. 11.. Elastic - The explanation must possess the intrinsic abilities to self organize, reorganize, give room to emerging order, accommodate new data comfortably, and react flexibly to attacks from within and from without.

In all these respects, Parapsychological explanations can qualify as scientific theories: they both satisfy most of the above conditions. But this apparent similarity is misleading.

Scientific theories must also be testable, verifiable, and refutable (falsifiable). The experiments that test their predictions must be repeatable and replicable in tightly controlled laboratory settings. All these elements are largely missing from Parapsychological "theories" and explanations. No experiment could be designed to test the statements within such explanations, to establish their truth-value and, thus, to convert them to theorems or hypotheses in a theory.

There are four reasons to account for this inability to test and prove (or falsify) Parapsychological theories:

1.. Ethical - To achieve results, subjects have to be ignorant of the reasons for experiments and their aims. Sometimes even the very fact that an experiment is taking place has to remain a secret (double blind experiments). Some experiments may involve unpleasant or even traumatic experiences. This is ethically unacceptable. 2.. The Psychological Uncertainty Principle - The initial state of a human subject in an experiment is usually fully established. But the very act of experimentation, the very processes of measurement and observation invariably influence and affect the participants and render this knowledge irrelevant. 3.. Uniqueness - Parapsychological experiments are, therefore, bound to be unique. They cannot be repeated or replicated elsewhere and at other times even when they are conducted with the SAME subjects (who are no longer the same owing to the effects of their participation). This is due to the aforementioned psychological uncertainty principle. Repeating the experiments with other subjects adversely affects the scientific value of the results. 4.. The undergeneration of testable hypotheses - Parapsychology does not generate a sufficient number of hypotheses, which can be subjected to scientific testing. This has to do with its fabulous (i.e., storytelling) nature. In a way, Parapsychology has affinity with some private languages. It is a form of art and, as such, is self-sufficient and self-contained. If structural, internal constraints are met, a statement is deemed true within the Parapsychology "canon" even if it does not satisfy external scientific requirements.