Turner at Sisteron: Further Observations

This article furthers the discussion begun in the article ‘Moonlight and its Implications: Turner at Sisteron’, posted on 8 November 2013. This follows a visit to the site on 12-14 April to verify the phenomenon depicted in a Turner watercolour of ‘Sisteron from the North-West’.

The effect in this watercolour is that of the setting sun casting the shadow of the citadel of Sisteron onto the facing rock wall of the Rocher de la Baume. Surprisingly this has never occasioned any notice, nor have I managed to discover any other artistic or even photographic representations. If readers know of any, or indeed have taken any themselves, I would be happy to see them.

Although it seems unlikely that Turner can be the only one to have ever noticed, the effect is not by any means a constant phenomenon at Sisteron. The shadow requires the sun to set in a specific part of the sky due west or very slightly either side. This occurs around the spring and autumn equinoxes, so mid-March to early April, or September to early October. The shadow is at its largest when the sun is at its lowest, and at its brightest just at the point before the disc makes contact with the horizon. Given that the western horizon at Sisteron is quite hilly, the optimal moment is when the sun sets in the dip immediately north of the Montagne de l’Ubac. Google Earth allows us to visualise the effect on 29 September.

Turner’s effect, however is more particular than even that, for it is conjoined with the nearly full moon rising in the south-east. When writing the first article I worked out using ‘Stellarium’ astronomical software that this conjunction might be observed around the 12th of April 2014. So with incautious trust in my calculations; still more in the likelihood of it being clear enough to see anything, flights were booked from Liverpool to Nice, car hire arranged and hotels reserved. My wife meanwhile looked upon the enterprise as implausibly eccentric and inevitably doomed.

So even she could see some justification for my astonishment when we arrived on site overlooking the Pont du Buech at 19.30 to find the sun sloping towards the western horizon and the nearly full moon rising in the gap between the Rocher de la Baume and the citadel of Sisteron, petty much exactly as shown by Turner. Still more so when fifteen minutes later the sun dipped low enough to raise the shadow of the citadel into view.

Almost as soon as the shadow became well defined, it began to fade. Within a minute it had gone. Click on any image below to enlarge and play sequence.

We were not the only ones about, but seemed to be the only ones with any interest in the event. We were the only ones with a camera and tripod, and I was the only one animatedly gesticulating.

The visit prompts a few ruminations. The first, and readers are bound to have noticed, is that my elation at photographing the effect appears exaggerated. The shadow, one might justifiably complain, is barely visible, and the moon hardly more than a dot. Enlarging the image helps a little, but quite plainly Turner shows both shadow and moon as more satisfactorily visible objects. In fact I would completely admit that the Turner is altogether more satisfactory a representation of the phenomenon than is the photograph. If one could perhaps enlarge the photograph so as to be as much in the scene as in reality then one might feel differently, but perception in reality is altogether more dynamic and active an experience than can be captured (except perhaps in singularities) by the photograph. So in reality the moon dominates the sky and the shadow the attention in exactly the way shown by Turner.

It seems hard to imagine that Turner spent much time in astronomical calculation, as diverting an activity as that assuredly is. On the other hand we may wonder that he just happened to come upon an effect that lasts barely ten minutes and only at certain times of year. It is possible that he was directed to it locally, but more likely that his routine method was such that being there at the right time, he was highly likely to discover it. We know that he visited Sisteron in late September. There is a sketchbook that records an itinerary from Genoa to Grenoble (TB CCXCV, labelled by Turner ‘Genoa to Grenoble’), and sketches at Nice earlier in the sequence (ff. 62a, 64) are dated 9 and 10 September. It was always his practice to look for special effects of light and shadow, particularly at dawn and sunset, and is was his regular routine to use all the hours of daylight – and sometimes of night – for work when was touring. It was also his regular habit to quarter the compass at his sites and to pay acute attention to the direction of light and its specificities. He might not have calculated such a phenomenon, but he was always abroad, alert and in position to witness any effects that might occur.

As much as we know that Turner made a journey via Sisteron in September, we do not yet know the year. Engagement with matters astronomical, might, however facilitate some speculation. We have already noted that the shadow of the citadel on the Rocher de la Baume is particular to a period around the equinoxes. We might next consider the position of the moon. The moon in Turner’s watercolour is lower than the summit of the Rocher de la Baume and in the south-east. For this to be the case, the time needs to be a few days before full, in its ‘gibbous’ (or brightening) phase. So assuming that Turner passed through Sisteron a couple of weeks after making his sketches in Nice, then we would be looking for a year when the full moon occurred towards the end of September or the first days of October. In the 1830s the years 1833 (28 Sept), 1836 (24th), 1838 (3 October) and 1839 (22 September) fit the bill; the later in the month the better. One of the dated sketches at Nice is more specific: It reads ‘Lunedi 10 September’ (TB CCXCV 64). Turner’s handwriting allows room for some uncertainty of reading, unfortunately, but Mondays fell on 10 September in 1832 and 1838. In the latter year the phase of the moon would have fitted perfectly with the shadow at its most optimal.

Note: Ian Warrell and I have chatted through the possibilities of dating this tour to Sisteron on several occasions, and considered 1838 as a possibility. I discounted 1836 in my exhibition catalogue Turner, Mont Blanc and the Val d’Aosta (Aosta 2000) on the grounds that the Fort Bard sketchbook clearly shows Turner head north from Turin via the Mont Cenis pass to Chambery. It seems highly unlikely that he would then have looped south from Chambery to Genoa. Various scholars have considered a possible link with the 1836 tour on the grounds that the style of the sketches and the work in watercolour is quite close. I would say that both pencil work and watercolour tends here to be freer, and a date of 1838 looks a very good fit in stylistic terms. There seems to be no impediment in terms of the documentation. A J Finberg (Life of J.M.W.Turner, 1939, p.370) established that ‘Turner was absent from all Academy meetings between August 6 and October 20’, which suggested that he was out of London at this time. Finberg surmised that he was at Margate, but this is based on the evidence of a dubious sighting on 23 September and a story that he witnessed the Fighting Temeraire being tugged to its last berth on 6 September. Neither contention has much solidity, and it seems a far more likely explanation for such a significant absence from London that he was making a continental trip as was his annual habit.

[Updated 30.12.2016, with colour reproduction of ‘Sisteron from the North West]