In The Turn of the Screw, Henry James pays tribute to many an author. Having based the governess on a choice of Austen characters, he develops her narrative along the lines of some classics from the previous century. The fact that the ghosts are human, for instance, while the governess perceives them as supernatural, reflects Anne Radcliffe’s The mysteries of Udolpho (1794). A novel that the governess mentions in chapter four. In the same sentence she recalls the mystery from Thornfield Hall in Jane Eyre (Charlotte Brontë ; 1847), from which novel James copied the autobiographical structure, the bad news from home, and the heroine’s apparent “tendency for deceit”. While the sentence itself effectively states that a plot along the lines of these novels would not have fooled her for long.
The screw’s final turn is a strand of poetry that I can’t discuss without violating the omerta. But it is definitely not written by Edgar Allan Poe ; the century’s undisputed maestro of gothic suspense, who Henry James seems to ignore alltogether. The strand of choice, meanwhile, is cunningly woven into the story’s prose. A method of wordsubstitution, so to speak, that leaves it virtually invisible to people who suffer from blindness for vital clues on a silver plate. But that also leaves the original poem virtually unchanged, because by this method even the slightest alteration of the original poem’s order of words is out of the question.
The introductory section is clearly inspired by Nathaniel Hawthorne’s Twice-Told Tales (1837). If only by its title. And its recollections of a an evening with friends place James in the position of the third narrator :
“Saturday, January 12th, 1895
Note here the ghost-story told me at Addington (evening of Thursday 10th), by the Archbishop of Canterbury: a mere vague, undetailed faint sketch of it – being all he had been told (very badly and imperfectly) by a lady who had no art of relation, and no clearness: the story of the young children (indefinite number and age) left to the care of servants in an old country-house, through the death, presumably, of parents. The servants, wicked and depraved, corrupt and deprave the children; the children are bad, full of evil, to a sinister degree. The servants die (the story vague about the way of it) and their apparitions, figures, return to haunt the house and children, to whom they seem to beckon, whom they invite and solicit, from across dangerous places, the deep ditch of a sunk fence, etc. – so that the children may destroy themselves, lose themselves by responding, by getting into their power. So long as the children are kept from them, they are not lost: but they try and try and try, these evil presences, to get hold of them. It is a question of the children ‘coming over to where they are’. It is all obscure and imperfect, the picture, the story, but there is a suggestion of strangely gruesome effect to it. The story to be told – tolerably obviously – by an outside spectator, observer.”
An author’s notebook is a sketchbook, and its notes are not penned down with the intention to share them with others. To question the honesty of Henry James in this entry is therefore in itself questionable. To be at the safe side, one better takes this entry as evidence of the source’s influence on the result, rather than the other way round. But that reversion looks rather temptatious :
location Addington Palace (1774) great old country-house
source ‘a lady’ a governess
told by Edward Benson (*1829) Douglas
date Thursday 10 Jan. 1895) Thursday 26 December
retold by Henry James (*1843) third narrator
Judged by the original story’s outlines, it is a good thing that Henry James left us a transcript of his own. The lack of quality, however, is worth noticing, because traditional folk tales are usually specific on details like names and places. As a result the original story looks very much like a hearsay account of a real incident. And that is exactly how Henry James presents it. Both in the 1908 preface, and in the introductory section to his transcript. And its narrator is an outsider. Yet, Henry James makes the heroine specific enough on names and places to suggest a traditional tale, which it can’t be without being specific on the exact nature of the danger that the apparitions represent. Yet, it is on that matter that the narrative goes evasive, while the story’s surviving version needs a very sharp observer indeed to recognize the children as bad. As a result, the original story’s specific-against-vague effect has been reversed, and a traditional folk tale is once again ruled out.
The status of the children is the key to all interpretation. And because our transcript version needs an Elizabeth Bennet to recognize them as bad, it left literary criticism sharply divided on the part played by the heroine. Is she responding on a genuine threat, or on a delusion? The answer is in James’s dealings with two pieces of evidence. One is the boy’s confession that he is ‘bad’. The other is the girl’s bad language. This evidence is in itself genuine enough, but it is placed in a context that raises questions about its reliability as proof of a supernatural presence. Not even the independent witness of the girl’s bad behaviour can chance that. A problem that in real life never fails to hamper attempts to verify the reported sighting of a ghost. As a result the sceptical ghosthunter still thinks himself fully entitled to doubt. Just as to be expected in cases when the reported ghost is actually there. What about that for deceitful ambiguity?
The last detail to notice, is that the first draft was made ‘much later’ in the sense that an author usually sleeps in the same room as his notebook. This is the only way to ensure that an interesting idea gets penned down before it fades from memory. This one didn’t fade at all, because the final transcript is exact on every point that can be checked : Edward Benson was old Trinity, and so is Douglas. The archbishop died within two years after telling the story. And, according to the upcoming dating attempt, so did Douglas. And he maintains the parallel by describing the governess as a true lady. If not by class, then at least by personality.