Using Google’s Ngram Viewer to explore the frequency and usage of the term “preternatural” in English language books published between 1600 and 1850 yields interesting results. Of course we should be cautious about drawing conclusions from these results for a variety of reasons (for one thing, this only searches books that have been digitized, and OCR is still difficult for early modern printed materials).
The word appeared most frequently in the second half of the seventeenth century. There were spikes again in the 1740s and 1780s. After 1800, the word preternatural was used much less often in print.
Before 1700, many writers were using preternatural to refer to phenomena associated with apparitions, witchcraft, magic, or demons. One example is Increase Mather’s An Essay for the Recording of Illustrious Providences (1684) in which he described “things praeternatural which have hapned in New-England” and included accounts of witches, demonic possession, and apparitions. After around the turn of the eighteenth century, the term loses those associations and was more frequently used in a medical context to indicate something unusual or idiopathic. The medical usage of preternature can also be found prior to 1700.
Samuel Johnson’s Dictionary of the English Language, published in 1755, defined preternature as “different from what is natural; irregular.” His examples of usage in quotations show no association with demons or magic at all. However, under the entry for “metaphysical” he noted an earlier meaning: “In Shakespeare it [metaphysical] means supernatural or preternatural.”
Another dictionary from this period, An Universal Etymological English Dictionary, defined preternatural as “beside or out of the Course of Nature, extraordinary.” In addition, it defined prodigy as “an Effect beyond Nature, a monstrous or preternatural Thing.”
A treatise published in 1823 by a medical doctor used preternatural with its older associations in An Essay on Apparitions, in which their appearance is accounted for by causes wholly independent of preternatural agency. The author acknowledged that ghost belief is universal, but argued that they had natural causes (such as hallucinations) rather than preternatural ones (such as being the returned spirits of the dead).
He wrote: “A celebrated conjuror, or mystic mason, with whom I had a conversation some years ago, told me, he could give me a receipt for a preparation of antimony, sulphur, &c., which, when burnt in a confined room, would so effect the person shut up in it, that he would fancy that he saw spectres and apparitions; and that, by throwing his voice into a particular part of the room, he could make the person believe he was holding converse with spirits” (Alderson, Essay on Apparitions, 46).