How to navigate a Viking longboat with a king, some bees and a DC-8

Feb 1, 2011 Alun Science, The Past 4 Comments atmospheric optics, History, imaging polarimetry, Medieval, Navigation, sky polarization, Viking navigation, Vikings

Vik­ing Ship. Photo (cc) Simon Bishop.

Jo Marchant has repor­ted on a new paper, On the trail of Vik­ings with polar­ized sky­light: exper­i­mental study of the atmo­spheric optical pre­requis­ites allow­ing polar­i­met­ric nav­ig­a­tion by Vik­ing sea­farers, for Nature news. She also adds more on her own blog includ­ing the link to the paper that you can access for free.

The research is part of an ongo­ing pro­ject by a col­lect­ive of sci­ent­ists to see if the Vik­ings could have nav­ig­ated the Atlantic in cloudy weather using polar­ised light viewed through crys­tals. There is no doubt that the Vik­ings were mas­ter nav­ig­at­ors, the dif­fi­culty is how did they navigate?

If the sun is vis­ible then they could have used a solar com­pass. This is a bit like a sun­dial. You make a wooden disc with a gnomon stick­ing up out of it. Then you scratch out a line show­ing where the edge of the gnomon shadow reaches dur­ing the day. Where it’s shortest is south and the oppos­ite dir­ec­tion is north. Now, when you go sail­ing the next day, you float it in a small tub of water — to make sure it’s hori­zontal — and then look at the shadow, when you turn the disc so the shadow matches the line you can work out where north is. It does mean know­ing morn­ing from after­noon, so that’s a prob­lem round noon if you’re not care­ful, but there error would not be huge. It wouldn’t work for weeks on end, because the path of the Sun would move in the sky, but it would be good for a month around mid­sum­mer. At least if there’s no cloud. That’s not always the case in the North Atlantic. Storms and fog are com­mon and if there’s neither of those then there’s often cloud.

Hor­váth et al. have been fol­low­ing the tale of the sun­stone, that gets a brief men­tion in a saga I can’t identify In the saga the Sun is obscured by cloud. Sig­urd iden­ti­fies the loc­a­tion of the sun by means unknown and Olaf, the king, uses a sun­stone by an unknown method to con­firm he’s right. I can­not tell you why it was neces­sary to nav­ig­ate imme­di­ately, nor what role the sun­stone plays in the saga. If it’s men­tioned in a way sim­ilar to sails or rig­ging, we could say that it’s prob­ably stand­ard equip­ment for sail­ing. On the other hand if the sun­stones are gifts from islanders then maybe check­ing a sun­stone is like recal­ib­rat­ing the sensor array in Star Trek, a mean­ing­less action that allows the story to move on without really resolv­ing any­thing. I don’t even know, from the bit of the saga quoted, if the sun­stone was used on a ship. There’s no real dis­cus­sion of his­tor­ical con­text, which is a con­cern for any­one inter­ested in Vik­ing nav­ig­a­tion for his­tor­ical reas­ons.

The appear­ance of the stone in the saga is one of three jus­ti­fic­a­tions for their use. The authors have oth­ers. At this point I implore you to read the art­icle your­self after this post, because this next bit must surely be over-simplifying their argument.

Hor­váth et al. say another reson for believ­ing sun­stones were used is that Scand­inavian pilots in DC-8s in the mid-20th cen­tury used a sky com­pass developed for the US Air Force to track the polar­isa­tion of light. How con­vin­cing you find this as an argu­ment depends on sim­ilar you link a Vik­ing crys­tal and a piece of 1940s sci­entific equip­ment is. The third argu­ment is simply too subtle for me to fol­low so I’ll quote it in full.

The third argu­ment for sky-polarimetric nav­ig­a­tion by Vik­ings is related with the dis­cov­ery of Frisch [4], that hon­ey­bees (Apis mel­lifera) are sens­it­ive to polar­ized light and use the dir­ec­tion of sky­light polar­iz­a­tion for nav­ig­a­tion, when the Sun is hid­den by clouds, but some clear regions of the sky are still vis­ible. A sim­ilar capa­city to detect polar­iz­a­tion pat­terns and ori­ent by means of sky polar­iz­a­tion was later found in many other arth­ro­pod spe­cies [5].

I can see how it shows polar­ised light is real, but I don’t think that’s dis­puted. To me say­ing bees can see polar­ised light there­fore Vik­ings used equip­ment to see polar­ised light is like say­ing bees can hover there­fore Vik­ings could have used hover packs to rise above the ship and see fur­ther into the dis­tance. I must be miss­ing some­thing pain­fully simple and obvi­ous, but I can’t see what. Please feel free to tell me what it is in the com­ments below. It’s baff­ling me.

What I liked about the paper was that bey­ond this, they also tackled some cri­ti­cisms of the sun­stone hypo­thesis. Roslund & Beck­man (1994) have argued that sun­stones aren’t neces­sary in partly cloud­free skies and that people could have loc­ated the Sun by obser­va­tion. The obvi­ous thing to do is some sort of a test and they refer back to a test by Barta, Hor­váth & Meyer-Rochow in 2005 to see if people could do this. They showed people fish-eye pho­tos on a mon­itor under vari­ous con­di­tions and asked people to say where they thought the Sun was like a game of Spot the Ball.

Unsur­pris­ingly they found it’s easier to see where the Sun is in clear skies. In the cloud­i­est skies the range of guesses var­ied more, which you’d also expect. They report the max­imum range of guesses var­ied between 8.1° and 162.9°.

That is unexpected.

That last res­ult, 162.9°, both­ers me a lot. I have no doubt it’s real and what they meas­ured. It’s pos­sible to do this on a mon­itor because you see all the sky ay once, but the sky is immers­ive. If you go out­side you’re unlikely to see the whole sky. You’ll see the dir­ec­tion you’re facing, but some it will be behind you.

It’s also tem­por­ally immers­ive. If you’re sail­ing then you’re under the sky from sun­rise to sun­set. You might see the sun in the morn­ing and if you do, you won’t be look­ing in that dir­ec­tion if you want to see the sun set in the even­ing. The pho­tos given to the test sub­jects were snap­shots, not sequences. This gives you more inform­a­tion if you want to see the dir­ec­tion of the Sun. If you’ve been out all day under the sky are you going to argue over whether the Sun is in the east or the west? That’s what an error of 162.9° implies to me.

It’s pos­sible that it’s a heavy cloud day in the morn­ing, with the chance of rain in the after­noon. You won’t see the Sun then, but your sun­stone might also have a prob­lem. The polar­isa­tion of light still hap­pens in cloudy or foggy days, but the light reach­ing your eye is reduced. You don’t just need polar­isa­tion to hap­pen, you need enough polar­ised light to see it by. The paper reports this isn’t likely in heavy over­cast skies, but it could work in fog — if the fog is dir­ectly lit by the Sun.

Jo Marchant’s blog post adds more to this with Susanne Åkesson say­ing that actu­ally the dif­fer­ence can be seen. To some extent there’s an argu­ment about how much light you need, but I’m will­ing to give Åkesson the bene­fit of the doubt on this. I don’t know if I could see it, but using one of these stone is not going to be a trivial skill. Nav­ig­at­ors would have much more exper­i­ence and be much more sens­it­ive to changes than someone pick­ing it up for the first time. How much dif­fer­ence this would make I can’t say because of lack of his­tor­ical evidence.

It’s this lack of his­tor­ical evid­ence that both­ers me. The paper is largely ahis­tor­ical, which is odd for a paper that really only has his­tor­ical rel­ev­ance. The pur­pose of the his­tory seems to be to jus­tify talk­ing about light polar­isa­tion, and some­times this simply doesn’t work.

The map of major routes, shows the Ice­land to Sval­bard and Ice­land to New­found­land routes as ‘main sail­ing routes’. It ignores the North Sea, which the Vik­ings reg­u­larly crossed and is prone to fog, per­fect sun­stone weather. There’s reason to believe that Sval­bard was vis­ited by the Vik­ings, though some ref­er­ences might be con­fused with Jan Mayen but I’d be amazed if it was one of the seven main routes sailed by Vik­ings. I can’t see what the source for this is or why it’s there apart from to draw long lines over the sea.

Hor­váth et al. have a dif­fi­cult task. There’s no archae­olo­gical evid­ence for sun­stone use and there might not be for many years. If you want to make a case that the stones were used in nav­ig­a­tion you’ll want to find one on a ship. The most likely place to find Vik­ing ships is at the bot­tom of the sea. Pro­spect­ing for ship­wrecks in the ocean is not going to be an easy in the forsee­able future.

Hav­ing read the paper I feel like I’m out of step with every­one else. While writ­ing this NewS­cient­ist have put up a story say­ing that there’s more evid­ence in favour of the sun­stone hypo­thesis. I think they could be used as a tool of last resort, like when a ship is fog­bound. I liked that they made a ser­i­ous effort to tackle cri­ti­cisms, but I found their dis­missal of Roslund & Beckman’s argu­ment against use under par­tially cloudy skies uncon­vin­cing, and it seems there’s doubts if they can work in fully over­cast con­di­tions. Given how his­tor­ic­ally weak the idea is, I’m not sure how rel­ev­ant the pos­sib­il­ity they could have been used is. Impossib­il­ity would rule sun­stones out, but there’s plenty of pos­sible things that could have happened that didn’t.

Read­ing the paper actu­ally made me less con­vinced about their use. I’ve gone from think­ing that it’s an inter­est­ing pro­posal to think­ing that even if it was pos­sible there’s no reason to believe sun­stones a reg­u­lar part of the nav­ig­a­tional equip­ment used on voy­ages. Maybe that’s why Sig­urd didn’t think to use one. Per­haps they were royal trinkets.

Update at midday:

Thanks to tweets from Jo Marchant (@JoMarchant) and Rebekah Hig­gitt (@beckyfh) I’m now mak­ing a bit more sense of this paper. My biggest mis­take has been to read this as if it’s a his­tory paper about what happened in the past. If you try read­ing it as using his­tory as a start­ing point to think about what might be, then it makes a bit more sense. Think of it as “A paper about the pos­sib­il­ity of using sun­stones for nav­ig­a­tion” with the Vik­ing saga as the reason to be inter­ested in the idea, and not the thing to be explained. In that light, “It’s not just Vik­ings that might have used polar­ised light to nav­ig­ate, DC-8s and bees do use it too.” isn’t quite so odd.

As a tightly focussed paper on could the Vik­ings have used sun­stones it works to a large extent. How­ever, could the Vik­ings have used sun­stones is a dif­fer­ent ques­tion to did the Vik­ings use sun­stones, which is what Roslund, Beck­man and most his­tor­i­ans are inter­ested in. I think it’s the blur­ring of the could and did ques­tions that derailed me. There simply isn’t the data in this paper to dis­cuss the did ques­tion. You can answer the could ques­tion with phys­ics and bio­logy, but did is going to require his­tory and archaeology.

Hor­vath, G., Barta, A., Pomozi, I., Suhai, B., Hegedus, R., Akesson, S., Meyer-Rochow, B., & Wehner, R. (2011). On the trail of Vik­ings with polar­ized sky­light: exper­i­mental study of the atmo­spheric optical pre­requis­ites allow­ing polar­i­met­ric nav­ig­a­tion by Vik­ing sea­farers Philo­soph­ical Trans­ac­tions of the Royal Soci­ety B: Bio­lo­gical Sci­ences, 366 (1565), 772–782 DOI: 10.1098/rstb.2010.0194

See also:

Barta, A., Hor­váth, G., & Meyer-Rochow, V. B. (2005). Psy­cho­phys­ical study of the visual sun loc­a­tion in pic­tures of cloudy and twi­light skies inspired by Vik­ing nav­ig­a­tion. Journal of the Optical Soci­ety of Amer­ica A, 22(6), 1023. OSA. DOI: 10.1364/JOSAA.22.001023

Roslund, C. & Beck­man, C., 1994. Dis­put­ing Vik­ing nav­ig­a­tion by polar­ized sky­light. Applied Optics, 33(21), p.4754–4755. DOI: 10.1364/AO.33.004754