The first page of Joseph W. Richards’ 1896 monograph on aluminium states that his work is dedicated to Friedrich Wöhler, "the discoverer of aluminium", and Henri Sainte-Claire Deville, founder of the aluminium industry in France. Richards, who wrote the book hoping it would help aluminium “[rank] next to iron in its usefulness to mankind”, clearly wanted to pay tribute to the men he saw as the fathers of aluminium: Wöhler, who was the first to successfully produce a small quantity of the hard-to-isolate metal in 1827, and Sainte-Claire Deville, who enabled its production on an industrial scale in 1854.
Figure 1 - Dedication of J. W. Richards' book Aluminium (1896, Philadelphia: H. C. Baird & Co.)
This sounds like a beautiful discovery story, but not everyone would agree with Richards’ version of it. According to the historian of chemistry J. R. Partington (1964, 4:323) for example, the Danish chemist Hans Christian Ørsted “should be regarded as an earlier discoverer” for his work on the metal in 1825. While researching the history of aluminium, I found that the alleged date of its discovery varied from 1754 to 1854, depending on the criteria of discovery that are used.
To make things even more confusing, the supposed discovery dates do not align with the first appearance of aluminium in a classification of elements, which was in a paper by Jöns Jakob Berzelius from 1811. Indeed, the addition of aluminium to lists of elements followed Humphry Davy’s work on alumina (aluminium oxide) and his coining of the name ‘aluminium’ in 1808 (though he first named it ‘alumium’). By 1815, aluminium appeared in most, if not all, classifications and textbooks: this shows it was thought to exist as an element long before its alleged discovery by Wöhler or Ørsted. In histories of aluminium, Davy’s work is most often mentioned as a failed attempt at isolating the metal and sometimes as having proved its existence – yet never as its discovery.
Figure 2 - Berzelius' classification of simple substances. Journal de Physique 53:279 (1811). You might wonder: why is it so difficult to identify the moment that aluminium was discovered and who should be credited for its discovery? Well, the case of aluminium actually isn’t exceptionally mysterious: the problem lies within the idea of scientific discovery itself. Most historians of science agree that scientific discoveries rarely, if ever, happen in a single moment, by a single person – even if that is how they are often presented in simplified accounts. Detailed historical studies of discoveries almost invariably show that they take place over a longer period of time and often involve multiple people.
There are different historical approaches that show that discoveries should be seen as a process rather than a clearly identifiable moment in time. For some, such as Thomas Kuhn for example, discoveries consist in multiple steps of observation and conceptualization that necessarily extend over a period of time: this requires “recognizing not only that something is but also what something is”. Others approach the complex nature of discovery by focusing on the social aspects of scientific practice. Simon Schaffer for instance, has described the process during which the story of an event is continually retold until it is established as a discovery in the form of a single observation or experiment carried out by a single scientist. In his view, the scientific community uses the retrospective label of ‘discovery’ to credit its most prized research practices.
The social and intellectual aspects of the complexity of discoveries are not incompatible, and both are relevant in the case of aluminium: its discovery required a series of experiments by multiple scientists over a long period of time, but it was retrospectively retold as a single discovery moment. In order to understand what happened, let’s go all the way back to the beginning.
Long before Wöhler’s, Ørsted’s and even Davy’s work on aluminium, there were suspicions that something like the ‘metal from alumina’ would exist. Alumina (aluminium oxide) had been known since the middle of the 18th century, as the base of a salt that could be found in clay. Because of its resemblance to salts that had a metallic base, it was suspected to have a similar composition. However, all 18th-century attempts to produce the metal from alumina were unsuccessful.
Between 1801 and 1812, Humphry Davy lectured at the Royal Institution in London. There, he used his voltaic pile to decompose many previously indecomposable substances, producing sodium, potassium, barium, strontium, calcium and magnesium in 1806 and 1807. This led Davy to believe that he might be able to decompose other, similar substances, such as alumina. He obtained some hopeful results but, in his own words, he was unable to obtain “decided evidences” in favour of the decomposition of alumina.
During the following years, most chemists accepted aluminium as an element despite the impossibility of producing it in metallic form. This was because of the strong resemblance between the chemical properties of alumina and those of barytes, strontia, lime and magnesia, which Davy had shown to be metallic oxides. It therefore made sense to classify alumina as a metallic oxide - and if alumina was a metallic oxide, the ‘metal from alumina’ (in other words, aluminium) had to be an element. The fact that aluminium could not be isolated could be explained by the hypothesis that it had a very high affinity for oxygen, which made it very difficult to decompose the oxide.
Attempts to produce aluminium continued during the next decades. In early 1825, Ørsted announced at the Danish Academy of Sciences that he had produced aluminium chloride and reduced it with the help of a potassium amalgam (a mixture of potassium and mercury). Despite appearing in German translation, Ørsted’s paper failed to attract much interest, and he did not continue his work on aluminium. In 1827 Wöhler visited Ørsted, who told him about his method for producing aluminium and encouraged him to investigate the metal further.
Back in Germany, Wöhler was unable to replicate Ørsted’s experiment, but he found that by replacing the potassium amalgam with pure potassium, he was able to obtain aluminium in the form of a grey powder. In his publication on the subject, Wöhler gave credit to Ørsted for discovering aluminium chloride but doubted the fact that his method produced pure aluminium. Ørsted wasn’t interested in obtaining priority for the isolation of aluminium, and it seems he did not contradict Wöhler on this point.
In 1854, Sainte-Claire Deville developed a new method to produce purer aluminium in larger quantities, which meant the metal could at last be used to produce objects. Deville, despite being encouraged by some to claim the discovery for himself, gave all honour to Wöhler, stating that “aluminium is known only thanks to Wöhler and his beautiful works”. He did not mention Davy, Ørsted, or any other earlier contributions.
Aluminium (at that time still more expensive than gold) was used to produce medals, one of which was sent to Wöhler, who became friends with Sainte-Claire Deville. In 1855, Wöhler and Sainte-Claire Deville were jointly awarded the Légion d’Honneur by Napoleon III. Throughout the 19th century, Wöhler was celebrated as the discoverer of aluminium. Aluminium was seen as the material of the future, and aluminium producers celebrated Wöhler as one of their heroes. Earlier works on aluminium were either forgotten or described as failed attempts at obtaining the metal.
This changed around 1921, when the Danish chemist J. Fogh reconstructed Ørsted’s experiments in order to show that pure aluminium could in fact be obtained using his method from 1825. He and his colleagues – possibly motivated by national pride – concluded that Ørsted deserved to be given priority as the discoverer of aluminium. They had considerable success and managed to effectively rewrite the history of aluminium, with the result that most 20th-century accounts of the discovery of aluminium credited Ørsted rather than Wöhler.
Figure 3 - An aluminium plaque bearing the face of Wöhler, probably produced by The Aluminium Company and sold during the1892 meeting of the American Chemical Society. Click here for its history .
Schaffer explains that ever since the late 18th century, scientific discovery has been seen as “an accomplishment and a prize”. This is true for the discovery of chemical elements as well, which was and still is extremely prestigious. The personal and national honour attached to the label of ‘element discoverer’ makes it difficult to make neutral statements about discovery. The history of aluminium is an example of how the label of discovery can serve different interests of the people involved, and the criteria of discovery can be modified in the process.
So is there a way to establish ‘objective’ criteria of discovery? Briefly put, there isn't. The identification of a discovery completely depends on the views of the people involved, and there is no way to judge what ‘really’ counts as a discovery above and beyond the discussions between such groups.
How then should we answer the question of who discovered aluminium? According to Helge Kragh, the question itself is misguided because it presupposes the existence of a clearly identifiable moment of discovery. The only way to answer it would be to point out the different steps that eventually led to the aluminium that we are all so familiar with today.
Whereas scientists and philosophers of science have long defended a view of discovery as the ‘eureka’-moment of a lone genius, one of the uses of history of science is to provide a more detailed description of the social and political context and the collaborative nature of scientific practice.
Figure 4 - Aluminium at the Exposition Universelle in Paris, 1937, by Roger Pérot. Copyright Institut de l'histoire de l'aluminium.