Thomas Kuhn describes two kinds of scientific inquiry in his book, "The Structure Of Scientific Revolutions." "Normal" science, according to Kuhn, is that body of work devoted to the support of particular theories. "Extraordinary" science is created by the problems left over by normal science, and in turn creates new paradigms.

Kuhn came to his conclusions about the nature of scientific inquiry by observing its history. Time and again, a deeply entrenched theory that had provided scientists with breakthroughs and answers would eventually prove inadequate. These theories would be replaced by newer ones, often being heavily resisted. The new theories would be hailed by proponants as The Answer, but without fail they would be replaced by yet newer and better theories.

Kuhn describes a cycle that is found throughout the history of science. Any theory starts out as a skeletal explanation for the most pressing problems left over by the previous theory. While the initial thrust of moving a new theory into acceptance is revolutionary, the work it leaves behind is "normal" science.

Throughout his book, Kuhn uses the word "paradigm" in a way that was new at the time.

Before Kuhn, "paradigm" meant a pattern that could be used to create sets, for example the paradigm for conjugating the verb "parler" in French allows us to conjugate other verbs using the same pattern.

Kuhn appropriated "paradigm" to mean something larger. In his writing "paradigm" means the predictive part of a theory that allows scientists to generate work and frames all observations in terms of the theory.

It's interesting to note that Kuhn published his book in 1962, and by the time of the ninth edition of the Oxford Concise Dictionary in 1995, his meaning for the word is the first entry, before the older definition it surpasses. This speaks to the influence this book has had.

Normal science sets about to "mop up" the unanswered questions created by the new, skeletal theory. This work involves conducting experiments that bolster the new theory and seek to enlarge the paradigm's range of applications beyond its initial, limited problem set. Kuhn even goes on to say, "No part of the aim of normal science is to call forth new sorts of phenomenon; indeed those that will not fit the box are often not seen at all." [Kuhn, pg 24]

Kuhn sets up in his explanation of normal science both its strengths and its weaknesses. In its favor, normal science provides us with the bulk of scientific discovery. It is the largest section of scientific work. Also, using a shared paradigm allows scientists a common ground, which they use to study their subjects with a greater amount of depth than they could without the paradigm. In the absence of a shared paradigm, scientists have to spend time recreating sets of basic assumptions with which to do work, and work done by different scientists are not commensurable.

In doing its own work, normal science uncovers counterinstances, real world phenomenon that contradicts the prevailing theory. At first normal science ignores and rationalizes these counterinstances. Left unresolved, however, the tension they create becomes great enough that scientists are motivated to look for a new and better theory, one that will add an explanation for the counterinstances without losing previously gained explanations. This is a time of "crisis" for science, the time when new theories are vying to usurp the older entrenched theory.

According to Kuhn, scientists do not reject theories simply because they do not account for all the evidence, or even when they are contradicted. "Once it has achieved the status of a paradigm, a scientific theory is declared invalid only if an alternate candidate is available to take its place." [Kuhn, pg 77] While counterinstances are a motivating force to go looking for other theories the accepted theory will not be replaced until a suitable candidate theory is discovered.

There are some interesting things to note about how the two modes of science differ. The defining difference between the two modes lies in the status of paradigms; in normal science one paradigm dominates and in extraordinary science new paradigms vie with the established one. From this difference we can say a few things about normal and extraordinary science.

In normal science, the assumptions that form the basis for the reigning paradigm are often treated as logical statements or even tautologies. Data that doesn't fit the paradigm is considered anomalous. Scientists work under the assumption that the anomalies will resolve themselves in light of other discoveries or the invention of new technologies.

But what makes a paradigm generate work for scientists lies in the fact that there are counterinstances. It is the work of normal science to extend the reach of the current paradigm as widely as possible. If the paradigm does become a simple logical statement, then its reach is, by definition, already infinite and no work needs to be done. What happens to these paradigms? They cease to be research problems and become engineering tools. [Kuhn, pg 79]

Normal science exists in a state of tension. Too many counterinstances, or a few select ones, and young turk scientists will seek new paradigms that work better. Not enough counterinstances and there is nothing to study.