ScienceWeek

ON ERRORS IN CHEMICAL THEORY

The non-scientific public often has a distorted view of the role of "errors" in science. In a closed system of thought, an admission or discovery of error can be of extreme significance, since an error in such a system can often cause the entire system to crash. Science, however, is an open system of thought, one in which errors, particularly errors of interpretation, are common. Indeed, an argument can be made that very often the progress of science depends on the discovery and correction of errors, so that the existence of errors is part of the fabric of science. Certainly, examples of important errors and their discovery and correction are numerous in the history of science. In general, a conceptual system without errors is dead -- there is nothing more to be learned.

In this context, the term "phlogiston" refers to a 17th and 18th century chemical theory involving a hypothetical principle of fire. The idea was that every combustible substance is in part composed of phlogiston, with the phenomenon of burning caused by the liberation of phlogiston and the "dephlogistonated" substance remaining as ash or residue. The phlogiston theory was experimentally discredited by Antoine Lavoisier (1743-1794) beginning in 1770, who showed that the newly discovered element oxygen was always involved in combustion, and by 1800 nearly every chemist recognized the correctness of Lavoisier's oxygen theory. One important exception was Joseph Priestley (1733-1804), the discoverer of oxygen and one of the greatest chemists of his time.

The phlogiston theory is an example of an erroneous theory that held sway for nearly 150 years until new facts and new analyses demolished it. The atomic theory of John Dalton (1766-1844) is an example of a theory erroneous in detail but correct in general and extremely fruitful in its effect on the scientific community. The chemical periodicity idea of John Newlands (1837-1898) is an example of an erroneous theory without much foundation but which anticipated a later correct theory. (In 1887, the UK Royal Society awarded Newlands the Davy medal for his anticipation 24 years earlier.)

The following points are made by Carmen J. Giunta (J. Chem. Ed. 5 May 01 78:623):

1) The author poses the question: Why did Priestley maintain the validity of the phlogiston theory in the face of the evidence and the weight of anti-phlogistic opinion? Not because he was in some way less objective or less open-minded than Lavoisier. As Priestley wrote to the leading French chemists at the close of the 18th century, "no man ought to surrender his own judgment to any mere /authority/, however respectable." The authority of the anti-phlogistic theory carried as little weight with Priestley near the end of his life as the authority of the phlogistic theory did when it was generally accepted -- or the authority of the established church or monarchy, for that matter. A freethinker politically and theologically, Priestley was caricatured in his day as "Dr. Phlogiston", a radical who trampled on beliefs held dear to church and state. A Unitarian minister, Priestley's religious ideas conflicted with those of the established Church of England. Politically, Priestley's sympathies for revolutionaries in America and then revolutionaries in France angered his neighbors, who destroyed his Birmingham home on Bastille Day 1791. Priestley left England and settled in the newly independent US, carrying his belief in phlogiston to his new home.

2) The author points out that John Dalton's atomic theory is an example of a set of mistakes that can be scientifically fruitful. Dalton's theory pictured chemical compounds much as they are pictured today, as atoms of different elements bound together, and the laws of definite proportions and multiple proportions follow naturally from such an atomistic view of chemical combination. But as valuable and fruitful as Dalton's work certainly was, it was equally certainly mistaken in several details. Dalton believed that atoms of the same element were identical. That all atoms of an element were identical was not simply a default position: Dalton considered the question to be an important one, and he concluded that if atoms differed in weight, the difference would have manifest itself in ways that were detectable in his time. With the discovery of isotopes in the early years of the 20th century, Dalton's idea of the identity of all atoms of the same element was proved an error. Similarly, Dalton's proposed formulas for molecules ("compound atoms") were also mistaken in many instances. The assumptions on which his formulas were based were arbitrary, with Dalton assuming the simplest possible formula if only one compound of a given pair of elements was known. Thus Dalton proposed HO for water and NH for ammonia.

3) Concerning the chemical periodicity idea of John Newlands, the author (Giunta) suggests this represents a classic case of "incoherent insight". Incoherent insights are, or contain, valid insights -- sometimes even important or useful insights. They also contain errors, inconsistencies, or other shortcomings. As a result, incoherent insights are usually not actively pursued and are not fruitful. Newlands published some short notes on the classification of the elements in which he pointed out several families of elements, the members of each group resembling each other in chemical properties, and their atomic weights also seeming to have a discernible pattern. Some of the groups had gaps, which Newlands predicted would be filled by elements not yet discovered. A few years later, Newlands listed all the elements then known in order of increasing atomic weight. When he did so, he noticed that chemical properties repeated with a period of 7 elements or a multiple of 7, and he thus formulated the "law of octaves". This law was not well received or influential, in contrast with the work of Mendeleev (1834-1907) on periodicity that appeared only a few years later. The author (Giunta) suggests that Newlands's work bore no fruit because its insight of chemical periodicity was not embedded in a coherent system that had either predictive or explanatory power.