Some Notes on Scientific Inquiry: the Role of Hypothesis

By Richard K. Moore, 22 November 1997

My activity on Internet can be characterized as an investigation into various questions, together with an investigation into techniques of effective presentation.

The net is a perfect medium for such pursuits—it is an Athenian Academy writ large, peopled with countless articulate philosophers and experts of every conceivable stripe, with varying levels of discernment, and espousing a wide range of traditional, popular, and innovative viewpoints. At each stage of an investigation one can find diverse critics ready and able to exhaustively dispute any current hypotheses or argument. This convenient and energetic forum enables rapid and economical refinment of ones ideas. Rebutters are my teachers, and reasoned refutations are valued; they may resent that I haven't read their particular recommended books, but if they're going to play on the net they shouldn't object to explaining in net-common terminology what may to them seem obvious.

Internet is uniquely potent in accelerating the development of hypotheses, especially when those hypotheses range over many traditional disciplines. And hypotheses are at the center of my formal investigative methods—methods, by the way, that are on a completely sound scientifc basis. My formal background is in mathematics (including logic), and math is the subject that most systematically develops the topic of “proof.” It was Euclid's geometry, for example, that first attempted to formalize the principles of deductive reasoning. Math doesn't really have any data; it only has its models and its proofs to play with. Philosophers may excel in analyzing such methods, but mathematicians are among the most proficient in applying the methods—it's like the difference between art historians and artists.

It was the facility of proof and argument that I took away from math; I never cared much about abelian groups or differential equations.

The first step in my method (and none of this is unique or original) is formally called INDUCTIVE HYPOTHESIS GENERATION. This is nothing more than guessing based on observation: What explanation seems to fit the facts? The tenative explanation might be original, or it might be borrowed, but if it is borrowed the authority of the source plays no further role—it is the hypothesis itself that one works with. Hypothesis generation is the process a police detective goes through after examining the initial evidence in a case.

Once an hypothesis is under consideration, investigation then proceeds along three threads in parallel. The first thread is to look for EVIDENCE for or against the hypothesis; the second thread is to EXPLORE the conditional consequences of the hypothesis (What would follow if it were true?); the third thread is to REFINE the statement and presentation of the hypothesis itself (like the detective developing the presentation for the prosecution). The three threads, it turns out, reinforce one another. The first thread is formally called “HYPOTHESIS TESTING”, the second “the ANALYTIC METHOD”, and the third “PUBLICATION” (:>) . Pursuit of these threads turns out to be highly productive: either the hypothesis is rapidly demolished by Internet detractors, which allows attention to be turned elsewhere, or else the hypothesis gains rapidly in refinement, substantance, and notoriety.

As example, some of my current hypotheses are:

(1) "The telecom system will rapidly evolve into a digital, high-bandwidth network, capable of delivering full two-way video to each user."
(2) "The mass-media industry will seek to use the telecomm network as its primary delivery channel as soon as such becomes operationally feasible."
(3) "The global capitalist elite have learned how to act coherently in their collective interests; they see those interests as being distinct from those of the nation-state; globalization has become the elite's vehicle for systematically furthering their interests."

These turn out to be very fruitful hypotheses: they are humble enough that thread one (validation) is not beyond the scope of reasonable endeavor (indeed many serious observers believe these hypotheses are already conclusively established), and yet they are bold enough that thread two (exploring consequences) leads to very significant and non-obvious observations, worth publishing in thread three.

To many who I find myself in debate with, the mere consideration of such hypotheses seems an unwarranted "leap". Such people have an impoverished scientific repertoire; they are imprisoned within the limitations of the deductive method: they want to consider proven facts and generally accepted beliefs only, and then they'll be happy to deduce and extrapolate conclusions from them.

Such people seem not to realize that most of the scientific facts and principles they now take for granted, and are comfortable in reasoning from, could never have been discovered by the deductive method alone, nor by sticking to accepted beliefs. Galileo, Darwin, Freud, Kepler, Descartes, Marx, Newton, and Einstein, to name a few, developed and employed the hypothesis-driven methodology I'm using, and I'm happy to borrow from them. They demonstrated conclusively that bold hypotheses, if astutely chosen and systematically pursued, can dramatically accelerate the advance of human understanding.

To flesh out the methodology just a bit more, let's look at hypothesis (2) re/ the mass-media and cyberspace. Note that without the hypothesis, there would be no reason to focus on the questions below; the value of the hypothesis lies in its ability to optimally focus the investigator's attention. Knee-jerk refusal to consider such hypothesis at all puts one in the position (and philosophical vintage) of the Scholastics who refused to look through Galileo's telescope.

Thread one, evidence-seeking, leads to such questions as:

How does the mass-media industry operate? Who owns it?
What is their business model? How deep are their pockets?
How do they conceptualize their marketplace?
How do they manage their distribution currently?
What are their market and product trends?
Where are they looking for future revenue growth?
How are they responding to new technologies?
How quickly do they learn and adapt?
What is their corporate culture? How pioneering are they?
What do we know about their actual plans and intentions?
What positions are they taking on regulatory issues?

The evidence to which these questions have led me turns out to be considerable, arugably conclusive. However it is not surprising that many serious Internet observers (on and off our lists) have not come to these same conclusions—that's simply because they haven't been examining the evidence with the right questions in mind. They may be more familiar with the overall room than I am, but they're overlooking a surprisingly significant corner, a corner which my hypothesis has guided me to examine closely.

Thread two, exploring consequences, indicates such questions as:

How does the mass-media business model map onto digital networking?
What would be their likely game-plan in pursing their business objectives in cyberspace?
What other players and game-plans that are likely to be involved?
What are the relative strengths of the players?
What are everyone's likely lobbying agendas?
How do those agendas fit with general directions in government policy?
How are the ownership and regulatory shakeouts likely to unfold?
What does that imply regarding the economic and regulatory regime likely to be adopted for cyberspace?
What uses can be anticipated from the technology itself, based on its ability to reliably and cost-effectively connect arbitrary points with high-bandwidth channels?

As challenging as this second set of questions appears to be, it is not really all that intractible. It has more variables and contingincies than the first set, but the players aren't that difficult to identify, their agendas are not deeply hidden, their initial game moves are already in evidence, and ample precedents exist in previous technological revolutions from which informed lessons can be drawn. Anticipation of a Big Brother, mass-media dominated cyberspace is not itself a “leap”, it is rather a well-reasoned most-likely outcome of the scenario indicated by the hypothesis. The only “leaps” are in the assumptions that cyberspace will be built, that the mass-media will jump in and play, and that neoliberal policies will continue to dominate politically.

If thread two is inadequately developed in the literature, that is only because the hypotheses has not been sufficiently entertained. Under the serious assumption of an aggressive mass-media policy, the fundamental cyberspace picture falls into place surprisingly readily. It would be a mistake to delay the difficult exploration thread until after the hypothesis is proven—that would be artificially serializing a learning process that can naturally proceed with productive paralellism. You might say an hypothesis has both CREDIBILITY and IMPORTANCE. Thread one seeks to establish credibility; thread two seeks to establish importance—and in the end thread three presents the two investigations together as one piece of work, which can then finally be offered even to the deductionist extrapolators amongst us.

For those who appreciate the method, and who can successfully switch their mental assumptions as they switch threads, all three threads are equally interesting and offer opportunities for productive participation.

But for those who can't appreciate the method, participation is typically limited to an unproductive finger which points at thread one: noting with dull repitition whatever doubts still remain regarding the base hypothesis.

Posted by Richard K. Moore
rkmoore@iol.ie
PO Box 26, Wexford, Ireland
www.iol.ie/~rkmoore/cyberjournal