Knowledge Gallery

Science and history seek to answer questions about the world around us. The theories and hypotheses used to interpret the evidence must be capable of rigorous testing and, if falsified, revised or rejected.1 Properly observed, this is a strength of science but also a limitation in that final proof or absolute fact are not the outcome of scientific work. There are other limitations to the extent by which information may be interpreted from the evidence and some of these are discussed here. Interpreting the evidence Both science and history seek to answer questions of “how? and “when?” in their respective fields. History also examines the evidence to answer questions of “why?” and “wherefore?” which may be in the form of human testimony or derived from the evidence itself. Cross section of the human knee showing the action of the cruciate (cruciform) ligaments as the knee is bent Science has difficulty over the questions of “why? and “wherefore?” because science only admits evidence of a material nature. Thus concepts of purpose, value, aesthetics, ethics and metaphysics are not derived from the evidence. Interpreting through a world-view The interpretative process can also be influenced by the knowledge, experience, philosophy and world-view of the interpreter. Gustaf Kossinna's interpretation of German prehistory was greatly influenced by theories to promote claims for an expanded German nation. This was much used and developed by the Nazi party in arguing the superiority of the Germanic peoples. Interpreting within the current paradigm Evidence is normally studied and presented from the viewpoint of the paradigm currently in vogue. This means that the expectations of a theory or paradigm colour perceptions in both research and interpretation. To quote Eldredge, 'new notions seldom arise from facts collected under the influence of old pictures of the world' 4 or as the editor of the Royal Institute of Philosophy stated, paraphrasing Popper, 'scientific consensus is sleep inducing'.5 In 1912 Alfred Wegener proposed the idea of continental drift suggesting that the continents once formed a single protocontinent which, over time had drifted apart to their current locations. The contemporary view, however, was of fixed continents and a rigid, shrinking earth. The American Association of Petroleum Geologists organized a symposium specifically to oppose his hypothesis.6 Fifty years later with additional evidence, his proposal came into its own with the development of the idea of plate tectonics. Interpretation and the "reinforcement syndrome" Hypotheses can be reinforced to a position of acceptability with the repeated addition of supporting data. This is particularly so in multi-disciplinary work and evidence from one discipline appears to confirm or give support to another. This is known as the “reinforcement syndrome”.7 A different aspect of the “reinforcement syndrome” concerns so-called publication bias. Confirmatory results will tend to be published prominently while non-confirmatory outcomes may be relegated to footnotes, obscure articles or not published at all. Negative evidence is still evidence.8 In the historical sciences – eg archaeology and geology - much of the evidence has disappeared. If there is nothing to observe, interpretation cannot proceed beyond speculation. Nor is it possible to test unique events.

Given the different approaches of scientific and historical enquiry, it is nevertheless from successfully tested hypotheses and confirmed observations and laws that theories about the world around us are developed. To provide a coherent story, the facts and observations with the supporting hypotheses and theories are brought together. For this purpose a model may be constructed. This is a hypothetical description or representation of the entity or process concerned and the interrelationships within it. These help to answer specific questions, find explanations and make comparisons or predictions.

In history and the historical sciences it is not possible to conduct a test on a past event. To test hypotheses certain assumptions may have to be made, for example that: present-day conditions applied at the time of the event; written or verbal records provide accurate evidence of the event; artefacts represent a certain type of behaviour; the event can be simulated by computer or other means. For an hypothesis and its assumptions to be validated, it is necessary to assess it against as many such assumptions as possible. A case study Consider the mysterious explosion that occurred near the Tunguska River in remote Siberia at 7.17 am on 30 June 1908. The evidence for the fact of the explosion is overwhelming: an estimated 80 million trees over 830 square miles (2,150 square kilometres) were knocked over as a result; seismic equipment in different parts of the world recorded an event equivalent to 5.0 on the Richter scale at this time; fluctuations in atmospheric pressure and effects on the light of the night sky were reported as far away as England; a deterioration of atmospheric transparency for several months was recorded in observatories in the US; the testimony of those near the event and press reports at the time. But what caused the explosion? The evidence for this has been elusive and in consequence many hypotheses have been developed. They range from an asteroid or comet impact to the airburst of a meteor just above the earth’s surface but no impact craters or meteorite debris have been found in the area. More speculative have been a UFO crash, a small black hole passing through the earth, the annihilation of antimatter from space, a natural nuclear explosion or, for those living in the area at the time, the beginning of the end of the world. Eye-witness accounts vary but include reference to a strong light in the sky, a hot wind, all the symptoms of an earthquake and sounds like heavy artillery fire over a period of time. Some press reports appear unreliable and one which reported they “were unable to study the meteorite closely because it was red hot...” quite untrue. The Universities of Bologna, Italy and Tomsk, Russia have been conducting field work at the site for a number of years.1 This has included a mapping of the fallen trees from which a trajectory azimuth and inclination of the cosmic body has been calculated. Based on this tree fall data and eyewitness accounts it is suggested that there was a multiple bolide which burnt up at a height of about 6-8 km above the ground.2 A key factor in the field work has been the search for any evidence of an impacting body. This included sampling from the bed of Lake Cheko, located close to the epicentre of the inferred event. A 3-D sonar image showed that the lake’s bed was funnel-shaped, quite unlike others in the area and is considered as possibly indicative of a meteoric impact. Sesmic-reflection profiles of the bottom of Lake Cheko revealed, beneath a finely layered recent sedimentation, a chaotic level and at about 10 metres below, a strong refective surface suggesting a high-density object. The unexpected discoveries about the bed of Lake Cheko have led to the development of a hypothesis that the lake is a water-filled impact crater. There is, however, no evidence that the lake is of 20th century origin and the degree of sedimentation supports this. The characteristic rim of an impact crater is missing.3 The hypothesis is currently being tested but little supporting evidence has been found to date; the results of recent magnetometer surveys and rock analyses have not yet been published.4 An alternative hypothesis assumes that the bright night skies observed as far away as England were the result of noctilucent clouds formed from the water vapour of a disintegrating comet.5 Other research involving supercomputers at the Sandia National Laboratories in the USA has simulated the fireball that might be expected from an asteroid exploding in the Earth’s atmosphere.This suggests that a small object would have been sufficient to cause the devastation of the Tunguska disaster.6 Thus the evidence supports the view that the Tunguska explosion was an extra-terrestrial event. Whether it was the outcome of a meteoric impact, a disintegrating comet or an asteroid explosion in the earth’s atmosphere remains uncertain.

As we observe the facts and behaviour so ideas form about their meaning. From these ideas we develop hypotheses to explain them. But are these ideas correct? A hypothesis, then, is a testable contribution to knowledge. In pure science it should be possible to repeat the same tests, conducted under the same conditions, and accurately predict the outcome each time. For an hypothesis and its assumptions to be validated, it is necessary to test different examples of the same occurrence in this way. Hypotheses The litmus test iillustrates an example of an idea derived from an observation. Arnau de Vilanova, a 13th century Spanish Alchemist discovered that a blue dye extracted from lichens turned red in contact with any acid. Would this always be so? This hypothesis has been well tested and validated since it was first discovered and is still used as a simple test of acidity today.