Applying the Scientific Method to Prehistory
What could be more scientific than the scientific method? A scientist observes an unexplained phenomenon. He or she gathers data, analyzes it, proposes a hypothesis to explain it, and tests it. The results are published in a peer-reviewed journal. Mission accomplished, right? Here are two papers on very different phenomena – one dealing with the geology of Mars, one dealing with DNA. Both papers follow the scientific method outlined above. Do they succeed in explaining the phenomena? If so, how trustworthy are the explanations?
- Mars strata. Regularly-spaced strata were photographed by the Mars Reconnaissance Orbiter at outcrops within the Becquerel crater on Mars. Six planetary scientists considered reasons why repeating strata 10 meters thick would be piled into hundred-meter bundles with a 10:1 ratio. “This repetition likely points to cyclicity in environmental conditions, possibly as a result of astronomical forcing,” they suggested in Science.1
What kinds of astronomical forces are available? The daily, yearly, and tidal cycles come to mind, but at Mars, there is also a tilting of the axis with a cycle of 100,000 years. This cycle is further modulated by 2.4-million-year influences. They chose the long-age cycles as the explanation for the sedimentary beds. “If deposition were forced by orbital variation, the rocks may have been deposited over tens of millions of years.” The conclusion was written up in a press release at Jet Propulsion Laboratory entitled, “NASA Orbiter Finds Martian Rock Record With 10 Beats to the Bar.”
Nevertheless, the team had to make some assumptions before proposing their explanation. They could not believe that 10-meter beds could accumulate in a year. “In contrast, deposition at orbital frequencies (~100,000 years) assumes a modest average accumulation rate of ~100 �m per year,” they said. “This value allows for alternating accumulation and erosion of sediment on shorter time scales, requiring only that the net deposition is roughly constant over long time scales.” They also had to assume that the obliquity cycle was relatively constant, “although the ancient history is unknown because of the chaotic nature of the obliquity over long time scales.” The pattern arose, the press release explained, through rhythmic variations in particle sizes due to changing winds as the climate varied by 10% each 100,000 year cycle, “or from how the particles were cemented together after deposition.”
With an explanation in hand, the scientists offered some predictions that might extend interpretation of Martian history:
The identification of quasi-periodic signals within these layered terrains provides a possible relative chronometer within the martian rock record. Orbital variations stand out as a possible driver of the observed quasi-periodicity, although definitive identification of the cycles involved will require additional information. Likewise, whereas an aeolian [wind-driven] scenario provides a clear link to orbital forcing, the specific formation model remains uncertain. Determination of formation time scales ultimately provides a calibration for interpreting the geological history of Mars. With the tentative but reasonable assumption that some water was required to lithify the Arabia deposits, the suggestion of orbital cyclicity implies that a hydrologic cycle may have been active at least intermittently over millions of years. In contrast to the catastrophic surface conditions inferred from impact craters and outflow channels, this strong cyclicity observed in the martian rock record depicts a fundamentally more predictable and regular environment in the ancient past.
- DNA introns: From the telescope to the microscope, we move to a scientific detective story involving DNA. One of the mysteries of the genome is why genes are interspersed with non-coding regions, called introns, that must be cut out of the messenger RNA after transcription by a complex machine called the spliceosome (09/12/2002, 09/17/2004).2 This phenomenon has stumped evolutionists for years (09/03/2003, 03/09/2006). A new paper by Catania and Lynch in PLoS Biology3 this week proposed a new hypothesis for the origin of introns.
The hypothesis is too complicated to describe, but relies on competition between various protein cofactors that assist in the transcription process. Genes are identified by start codons and stop codons. If a premature termination codon (PTC) becomes inserted in a gene, rendering it non-functional, the cofactors and proofreading machines, along with natural selection, may cause introns to grow on both sides of it. Introns result as an artifact of “crosstalk” between these factors. A gene with a new intron can still remain active, even if its RNA transcripts are discarded by nonsense-mediated decay (NMD), a proofreading process. “Such an allele can then be subject to positive selection for subsequent mutations that improve splicing of the modified region.” Once the allele’s transcripts survive NMD, they can still be selected if the protein product retains some function. As a result of this “intronization” process, they predict new introns will be short, and multiples of three, to preserve the reading frame. “Unless excision of the newly intronized coding sequence has sufficiently large deleterious consequences,” they proposed, “the fixation of the novel intron may be either selectively neutral or promoted by natural selection.” Their prediction of short introns in multiples of three appears to be borne out in six different eukaryote genomes they checked.
Simple as this proposal seems, there are many complications. Not all eukaryote genomes contain introns, and those that do have widely varying numbers of them. In addition, there seem to be highly-conserved introns in non-coding regions of the genome (05/27/2004, 08/18/2007, 10/08/2008). A scientific hypothesis has to be adequate for the exceptions as well as the rule. They discussed how non-coding regions might gain introns: either they were once coding regions, or suffered from upstream premature start codons, or both. As for introns in non-coding RNA genes, “it can be postulated that fortuitous endogenous events may on rare occasions promote splicing in noncoding RNAs, in such a way as to prevent more harmful secondary structures.” And as for eukaryote genes lacking introns, perhaps they were lost at some point. Or, perhaps they play a role in nuclear export of transcripts, and so are resistant to intron gain. “Although it remains to be proven, it is possible that the relative abundance of these elements that inhibit splicing plays a role in establishing different levels of intron-richness between eukaryotic species,” they said. Two different species of yeast, for instance, differ from 470 introns in one to 4,600 in another. Fruit flies have 38,000; humans, 140,000. They admitted early on that “Explaining the causes and functional implications of this uneven distribution requires understanding why spliceosomal introns exist in the first place and what the evolutionary origin(s) of these sequences are—a problem that has proved a conundrum for the past 30 years.”
In conclusion, they felt their “novel hypothesis” was at least a good start in explaining this puzzle:
Despite the mutational hazard associated with intron presence and proliferation, we argue that, at least initially, introns might represent a favorable life line for an allele that has acquired an ORF-disrupting mutation. In this sense, in-frame stop codons need not be dead ends, as often believed, but rather sequences that occasionally facilitate the evolution of eukaryotic gene structure, possibly favoring not only intronization, but also processes such as exonization (following a PTC loss). Further experimental validation of our hypothesis would not only support the idea that intron birth/death rates depend on both the population-genetic and the intracellular environment, but also shed light on a surprising aspect of the evolution of eukaryotic gene structure, i.e., the ongoing, stochastic process of mutual conversion between exons and introns within genes.
Two papers selected from the science journals. Though they deal with vastly different phenomena, they have several things in common. They deal with singular prehistoric processes not subject to the usual scientific requirements of repeatability, observation and testability: i.e., even if aspects of the phenomena can be seen today or repeated in a lab, that would provide no guarantee that the gross phenomena were produced that way in the unobservable past. In addition, the papers can be considered representative of today’s scientific approach to explaining natural puzzles that are presumably the result of natural causes acting over vast aeons of time. The only data available for study in such cases is the collective effect of multiple causes that may, in some combination, have acted in the past. This educated guesswork is what we perceive as normal science.
1. Lewis et al, “Quasi-Periodic Bedding in the Sedimentary Rock Record of Mars,” Science, 5 December 2008: Vol. 322. no. 5907, pp. 1532-1535, DOI: 10.1126/science.1161870.
2. Prokaryotes have “Group I” introns that are self-splicing; eukaryotes have “Group II” introns that are spliced by the spliceosome.
3. Francesco Catania and Michael Lynch, “Where Do Introns Come From?”, Public Library of Science: Biology, Vol. 6, No. 11, e283 doi:10.1371/journal.pbio.0060283.
“We don’t know enough about the unknown to know that it is unknowable.” – G.K. Chesterton.
“It isn’t what we don’t know that gives us trouble, it’s what we know that ain’t so.” – Will Rogers.
To the casual observer, these two papers represent noble activities of scientists seeking understanding. We respect their abilities and in no way denigrate their efforts, nor fail to honor their scholarship. It takes years of education, training, and experience to become knowledgeable enough to write on such subjects. They associate with scholarly individuals. They employ state-of-the-art equipment to gather the observations. Nevertheless, we need to ask serious questions before just trusting their conclusions. Namely: are their theories true? If their theories are mere stepping-stones on the way to a more complete understanding, how far along does their work bring us? How much farther is there to go? Is it possible we will never know the answers? If so, why should we have any confidence that partial answers are better than none, given the possibility that partial answers can be flat wrong? And is the ability to gain understanding of nature limited to scientists? Are the same methods, or different ones, potentially just as valid for non-scientists to employ in the search for understanding? The Science Academy will always validate itself. Its confident claims need to be cross-examined by the prosecution (philosophers of science) and adjudicated by a jury of their peers (fellow human beings), especially when they foot the bill.
Thomas Kuhn, whose 1961 book The Structure of Scientific Revolutions upset the apple cart of the perceived authority of science, described “normal science” as a puzzle-solving activity of workers within a paradigm. These workers tend to be satisfied with the paradigm (which can be described as a presuppositional framework that defines what constitutes good science and what puzzles deserve to be investigated). Scientists tackle puzzles that are assumed to have answers within the paradigm; they are not motivated to challenge the paradigm itself, Kuhn taught. As such, the science community can be likened to a guild of interested parties who cheer one another along but marginalize others outside the paradigm. Kuhn even suggested that scientists are incapable of understanding other paradigms, because they speak a different language: for example, a Newtonian means one thing by mass, but an Einsteinian means something quite different.
The current feeling of many scholars is that Kuhn may have oversimplified things, but his ideas cannot be dismissed. The Kuhnian Revolution spawned related fields like Sociology of Science, History of Science, and Rhetoric of Science – fields now enjoying their own academic departments at major universities. These departments turned the scientific method on science itself. They pulled the plug on the triumphal parade of science as an inexorable March of Progress toward The Truth. Science now had to be treated like any other enterprise of fallible human beings. Philosophers, sociologists, historians and rhetoricians sliced and diced science into little bits. What do we mean by scientific discovery? What do we mean by a scientific explanation? What branches of science should be included in the science department–political science? economic science? science of mind? The questions led to further attacks on the presumptive authority of science. Prediction and falsification were thrown out as reliable indicators of scientific validity. The Humanities departments rose up to dethrone the Science department. Sociologists wrote papers and books on the way scientists “manufacture” reality; they analyzed the social and emotional factors that motivate them, and questioned the validity of their claims. Postmodernism was born, as influential sociologists portrayed science as a mere text, subject to a number of equally-valid interpretations.
The scientists struck back in the Science Wars of the 1990s and largely succeeded, more by force of rhetoric and endurance than by winning the debate. Many scientists today recline comfortably in a posture called “scientific realism.” Though more restrained in its epistemic reach than the now-discredited Logical Positivism of the 1930s, it still asserts to have reliable contact with Reality. But scientific realism is still devilishly hard to defend in the post-Kuhn world. Scientists presume their work relates to nature as it is, and so they tend to work like the positivists did. They carry some of the same positivist baggage into their work: such as the assumption that everything in nature is the province of Science, and Science is better than other modes of inquiry. (Don’t ask them what “Science” means, though, because no one has been able to define demarcation criteria between science and pseudoscience that keeps the good stuff in and the bad stuff out; nor can they describe a foolproof “scientific method” that is unique to the Science Department.) In 2008, meanwhile, Science marches on, funded largely by the federal government, hoping the public has forgotten (or never heard of) the deep controversies over the nature of science.
All that was a necessary prelude to asking questions about these two papers. These papers are, in a sense, a call to respect by fellow human beings. They want you, the reader, to acknowledge that the authors have hit on something approaching “the truth” about reality. They expect you to assume that their explanations are valid for these phenomena whose effects are observable today, though the historical sequence is not. We are to respect their opinions and speculations because they are, after all, scientists, and had to work hard and learn a lot to earn that title. We should give honor to whom honor is due. Honor, however, does not necessarily correlate with truth.
We should notice first off that the puzzles they examined are not traditional scientific puzzles. Faraday or Joule could repeat experiments on magnets and energy over and over and over. Their experiments could be replicated by others. The laws they described led to inventions – motors, space heaters, electromagnets – that validate their conclusions every day. How, though, is one supposed to recreate and observe the history of Mars for supposed millions of years? How are we to replicate the convoluted history of intronization, or know the time scale over which changes took place? We see Mars now; we see introns now, but we don’t see what happened before. Even the tiny timeslice available to us to watch present processes at work provides no guarantee other factors were not involved in the past. The practicality of their opinions, furthermore, seems dubious.
The scientist at this point will appeal to other factors to buttress the authority of the “scientific” approach expressed in these papers: the vast corpus of published work by other scientists, the track record of science, the collective expertise (and hence authority) of the scientific community, the value of peer review, the quality of their data, and the perceived value of their efforts as measured by the willingness of the government to fund their work, to name a few. These factors may be fine for observable, repeatable, testable things in the present, but ask yourself if they really guarantee reliability for inferences about the unobservable past. Remember that an obvious inference can be wrong. Imagine a group of scientists searching for the prince in Mark Twain’s The Prince and the Pauper. They might spend most of their time interviewing candidates in the castle. One might publish a detailed hypothesis about why the son of some nobleman is the best candidate. Meanwhile, the real prince was out hobnobbing with beggars on the street.
CEH wishes to help scientists and observers of science respect good scholarship, value knowledge and understanding, and appreciate the wonders of nature without treating the Science Academy as a priesthood. These papers may have gotten their explanations right. They could be very wrong. When evaluating a scientific hypothesis or explanation, be aware of the following factors that have nothing to do with the truth of the explanation.
- Hidden assumptions: The Mars paper treated billions of years as a given. The intron paper treated evolution as a given. Neither assumption is a prerequisite for explaining the phenomenon.
- Social pressure: The maverick scientist is largely a myth. Most scientists attend regular conferences with colleagues in their field. Human desires for respect and recognition, and avoidance of being ostracized, can be powerful. Peers can be tolerant of your being unconventional—to a point. Some will think to the corners of a box; few may be willing to think outside the box.
- Momentum: The force of tradition can be powerful, even in the sciences. Civil engineers usually build onto and adapt existing infrastructure (e.g., primary road patterns) rather than tearing down a city and starting over. In the same way, the geologic column and the evolutionary tree of life are unlikely to be replaced just because of some contrary data or the wishes of a maverick. It’s too hard to start over. Think of all the books and papers that would be obsolete. These factors tend to force thinking along certain paths irrespective of their validity.
- Personalities: Certain champions in science tend to garner a following like philosophers did in ancient Athens. Their respect may be more due to their rhetorical skill than their contact with Reality.
- Incompleteness: We saw in the 10/16/2008 entry that science is incapable of validating itself.
- Missing or misleading data: You can never know what critical data is missing that would change the interpretation drastically. Consider historians studying ancient Rome. Without the text of Mark Antony’s speech at Julius Caesar’s funeral, or a recording of it, they can only infer what he said from the effects it had on the audience. Shakespeare’s version may feel like a reasonable facsimile, but does it reflect what Antony actually said? Who could possibly tell?
- Misdirection: The intron paper did not concern itself with the origin of genetic information and the elaborate machinery that translates it. Those are arguably much more interesting questions. The neo-Darwinist paradigm has already ruled on such matters, attributing any and all genetic information to mutation and natural selection. The paradigm not only sets the agenda, it marginalizes all who doubt the creative powers of natural selection: e.g., those in the Intelligent Design movement, who are routinely criticized as being “outside of science” because they don’t ask The Right Questions.
- Auxiliary hypotheses: Both papers required auxiliary hypotheses to buttress the main hypothesis. The Mars paper referred to climactic and fluvial factors that might have cemented the layers, but then turned and used the main hypothesis to inform the auxiliary hypotheses (i.e., the climate history of Mars). The intron paper leaned on factors that might provide immunity to intronization. It also expected that mutations and selection would reactivate genes silenced by introns, and postulated that large non-coding regions must have been genes in the past. How many buttresses does it take before a reassessment of the soundness of the central edifice is called for? Who makes the call?
- Myth of progress: Scientists often assume that knowledge is progressive and cumulative. The observations may get more detailed, but the paradigm could be progressing only in the details. Think of the dining and decorating getting better and better on a train headed the wrong way.
- Mental pictures: Scientists, like other humans, are subject to envisioning the world according to personal preferences. Their mental pictures of how the world came to be, and how it operates, can bias their research and their approach to doing research.
- Compartmentalization: We tend to think of “Science” as a department unto itself. Actually, its linkages to history, law, economics, philosophy, psychology, theology, rhetoric, and aesthetics are strong. A corollary is that scholars in each of these other departments employ reasoning similar to that of the scientist.
- Segregation: Scientists tend to live in their own communes, working the fields for their mutual benefit. If a scientist advances his or her department, gets more funding, wins a Nobel Prize or other recognition, that is considered a success—whether or not his findings are true.
- Integration: Once you are inaugurated into the ranks of The Scientist Guild (perhaps by earning a PhD, getting your first paper published, or getting hired at a research institution) is everything you do scientific from then on? Does being a scientist make you an Authority? We saw in the 10/21/2008 entry that it might just turn you into an insufferable know-it-all.
- Fallibility of peer review: The reviewers of a paper are just as human as the authors. They are subject to the same biases and social pressures as the rest of us. Criticisms of peer review have been growing in recent years: anonymous reviewers may reject a competitor’s paper; reviewers may be unwilling to validate radical departures from the paradigm; and reviewers may be unable or willing to check the facts sufficiently to prevent fraud.
- Fallibility of references: The Mars paper contained 32 references; the intron paper, 137. This tends to impress readers and win credibility points. We saw in the 3/17/2006 entry, however, that scientific journals can perpetuate bad ideas. Authors tend to reference earlier papers as appeals to authority as surely as Medieval scholars referenced Aristotle. Researchers lack the time to thoroughly read all those references, much less replicate their results. References, therefore, valuable as they are to any scholar, can contribute to chains of reasoning that, true or not, become so strong they are hard to break.
- Peer pressure: “Publish or perish” and other social pressures (the desire to get published in high-profile journals or get listed as a co-author among respected authorities) might bias one’s ability to think clearly and independently about the phenomenon in question.
- The money trail: “Follow the money” is good advice in science as well as politics. Big money is supporting research on global warming, embryonic stem cells and certain forms of cancer to the exclusion of other maladies. Funding has no necessary connection to the validity or importance of the subject matter, but can have profound effects on the motivation of scientists. When the NSF funds millions of dollars to study astrobiology or Darwin’s tree of life, do you think the recipients are going to come back and report that the whole exercise was a dead end?
- Priority of the paradigm: No one comes to a natural puzzle with a mind like a blank slate. All the influences above contribute to approaching the puzzle with a bias. The true solution to the puzzle could be very different from the solution the consensus is working on (see 05/01/2008 commentary). For a dramatic example, read the quotes on a page from the US Senate Committee on Environment and Public Works about global warming. CEH does not take a position on global warming. But just imagine what colossal impact a collapse of the human-caused global-warming consensus could have on the public perception of science, considering that Al Gore won a Nobel Prize on the subject and the world is poised to take drastic economic measures because of the consensus, which appears to be jealously guarded by an institution (the IPCC); read the hysteria in a BBC News article and look for the word “progress.”
So what are we suggesting? Toss out these papers as worthless speculation? No: you do what any good scholar should do. Apply critical thinking skills. Follow the money. Question assumptions. Look for hidden biases and conflicts of interest. Define the terms. Understand the context. Ask the right questions. Separate observation from interpretation. Respect the limits of knowledge. Hold judgment tentatively, realizing that scientific revolutions happen. Identify your authorities. Prove all things. Abhor what is evil; cling to what is good. Know them by their fruits. Whatever is good, honorable, true and of good report, think on these things. You didn’t learn that in science class. You learned it in the Bible.