Constellations in the art of Leonardo da Vinci

[Robert Tulip]

The ancient practice of imagining pictures in the patterns of the stars appears to have inspired Leonardo da Vinci in one of his famous paintings. The constellations along the ecliptic bear a striking comparison to the men either side of the central figure in this painting, as shown in the attached.

The poses of all the characters in the painting have an apparent correspondence with the stars of their respective.constellations, but some are more precise than others.

I would be grateful for comments on whether the attached matching between characters and stars appears to be accurate.

Last Supper Zodiac.jpg

[Jens]

Maybe I lack imagination, but I looked at it and it didn't look that way to me. I would think that if Leonardo has wanted to make that analogy, he could have done so through other subtle clues.

[Robert Tulip]

Thanks Jens, it is obvious to me, but I know the clues to look for. I have provided here more detailed depictions with overlays of the stars of each constellation against the spot where they are used as the template in the painting. I will also follow up by showing how these patterns are seen in the night sky looking south from Florence.

[Strange]

Can't see it myself, I'm afarid. At first I saw a rough correspondence in the first figures on the left and right - but then realized that is because they are both leaning outwards and the image of the constellations are tilted the same way. Your more detailed version doesn't really work for me. There are some rough coincidences between hands and stars but no consistency between how they are applied. For example, the middle picture in the second row has a V of stars lined up with a hand and a "column" of stars lined up with a finger. But the one on the left has stars arbitrarily positioned on the hands, while in the one on the right, the stars are spread out below the hands.

And then some of them are matched with heads, arms or overall body position. I imagine that if he had wanted to put this information in the picture it would have been far less ambiguous.

I am quite sure that after shuffling the constellations (or using made up patterns) one could create equally good matches.

[Robert Tulip]

The most unambiguous correlation is between the dagger of Judas and the sting of the scorpion. It was seeing this that made me look at the rest of the figures, to see how the shape of their respective stars form a template.

From Wikipedia, we learn that Leonardo habitually used mirror writing, from right to left. Indeed, the famous Vitruvian Man used this backwards script. Wikipedia provides the following intriguing commentary on the Vitruvian Man, which I suggest could apply also to The Last Supper:

This image exemplifies the blend of art and science during the Renaissance and provides the perfect example of Leonardo's keen interest in proportion. In addition, this picture represents a cornerstone of Leonardo's attempts to relate man to nature. Encyclopaedia Britannica online states, "Leonardo envisaged the great picture chart of the human body he had produced through his anatomical drawings and Vitruvian Man as a cosmography of the microcosm. He believed the workings of the human body to be an analogy for the workings of the universe." According to Leonardo's preview in the accompanying text, written in mirror writing, it was made as a study of the proportions of the human body

As a 'cosmography', and an 'analogy for the workings of the universe', the use of constellations as a template for The Last Supper could well make sense. If so, the concealment of this source adds a further layer of astronomical meaning to this famous work.

Leonardo had a keen interest in astronomy. In his Notebooks, he comments "The earth is not in the centre of the Sun’s orbit nor at the centre of the universe, but in the centre of its companion elements, and united with them." What he meant by 'companion elements' is unclear, but I would hesitate before writing it off as just pre-modern or superstitious. One possibility, just speculating, may be that he means the earth is at the centre of its seasons, marked by the annual passage of the stars.

My claim here is purely that Leonardo used the stars that the sun passes in its annual journey as his template, while stretching and flipping them to conceal his intent, but definitely keeping the same order as observed in the sky. These stars, viewed in the night sky from Italy, appear in the south in the order shown in the painting. The attached diagrams, using SkyGazer astronomy software, show the path of the sun as a yellow line through the twelve signs along the ecliptic. I have flipped and stretched each constellation to match the templates of the painting. Far from being arbitrary like patterns in clouds, the exact order shown here has the appearance of being precise and deliberate.

Signs of the Zodiac as Template for The Last Supper.gif

[Tog]

The ones I can see fitting, are different than the way you depict them. Rather than the hands, I was looking at the overall body positions, but I have Sagittarius and Scorpius reversed. The way I fit them in also fits a lot better in my mind than your second image with the overlays.

I'm going to have to cast a vote for pareidolia. We're all seeing it the way we think is right.

Also, I also thought Judas was the one with the dark hair holding the bag.

[grapes]

The most unambiguous correlation is between the dagger of Judas and the sting of the scorpion.

Isn't Peter the one with the knife?

One possibility, just speculating, may be that he means the earth is at the centre of its seasons, marked by the annual passage of the stars.

What do you mean by that?

My claim here is purely that Leonardo used the stars that the sun passes in its annual journey as his template, while stretching and flipping them to conceal his intent, but definitely keeping the same order as observed in the sky. These stars, viewed in the night sky from Italy, appear in the south in the order shown in the painting.

The order is not unique to Italy. It's the same everywhere (adjusted for standing on one's head).

[Robert Tulip]

Judas was the one with the dark hair holding the bag.

Yes, you are right. I read a description which listed the characters from left to right and thought the order was heads not bodies. The guy with the dagger/sword is Peter.

I don't think this is pareidolia, like images in clouds. How I see it is that Leonardo deliberately set out to conceal depictions of the stars in the painting. While he stretched and reversed the template in some cases, like in his mirror writing, he kept the same order as we see in the sky. For each of the twelve constellations of the ecliptic, you can look to see if it is concealed in the position matching the natural order, and the answer for all twelve is yes.

Here is how I see it, from right to left.

Aries: three main stars with long line and short line joined at angle of 150 degrees match to hands
Taurus: One hand forms the triangle of the Hyades around Aldebaran, the other hand forms the Pleiades
Gemini: Two arms in parallel lines ending in the twins Castor and Pollux
Cancer: Head and arms form 'steering wheel' in same shape as stars
Leo: left hand cupped like lion's head (Regulus), right hand stretched out as tail (Denebola)
Virgo: single hand similar to shape of stars

Christ: Pisces (equinox point) with right hand as circlet, head as knot and left hand as second fish

Libra: two hands together at point of the balance with arms matching the weights
Scorpius: Peter as match to shape of scorpion with head as head, arm as body and sword as sting
Sagittarius: Judas holding bag with arms and hands matching shape of stars
Capricorn: two hands and head form triangle
Aquarius: body and left hand match stars, with left arm stretched out
Pisces: arms and hands match to two lines of the fishes joined by knot at shoulder

[Robert Tulip]

Isn't Peter the one with the knife?

Yes, see last post. Apologies.

What do you mean by that?

Leonardo was saying, as I read him, that while it is true that the earth orbits the sun, there is still a cultural meaning in regarding the earth as the center, with the apparent position of the sun against the stars marking the seasons. That is a bit off topic so happy to drop it. I only mentioned it as evidence that Leonardo had an interest in astronomy and even in 1500 understood that the solar system is basically heliocentric.

The order is not unique to Italy. It's the same everywhere (adjusted for standing on one's head).

Actually, living in Australia myself, we see the ecliptic and the sun moving anti-clockwise across the north, so we see the stars in reverse order compared to the northern hemisphere, ie Taurus appears to the right of Aries etc.

[Robert Tulip]

The ones I can see fitting, are different than the way you depict them. Rather than the hands, I was looking at the overall body positions, but I have Sagittarius and Scorpius reversed. The way I fit them in also fits a lot better in my mind than your second image with the overlays. I'm going to have to cast a vote for pareidolia. We're all seeing it the way we think is right. Also, I also thought Judas was the one with the dark hair holding the bag.

To respond to the suggestion that the zodiac template was not actually used by Leonardo, I have made the attached diagram showing equivalent distortions of the first twelve letters of the alphabet. The chance of such a structured pattern occurring in the same order as the stars in the sky is miniscule.
How Leonardo used the signs as templates.GIF

[grapes]

To respond to the suggestion that the zodiac template was not actually used by Leonardo, I have made the attached diagram showing equivalent distortions of the first twelve letters of the alphabet. The chance of such a structured pattern occurring in the same order as the stars in the sky is miniscule.
How Leonardo used the signs as templates.GIF

Your comment there, that it's 12 to the 12th, seems to imply that there is a 1/12 probability of each correspondence, raised to the 12th power. Is that it?

But as near as I can see, if you allow complete freedom to distort like that, it's not 1/12, but it's near certainty that you'll be able to fit--that is, it's near certainty that you can fit any random pattern that way.

[Jens]

To respond to the suggestion that the zodiac template was not actually used by Leonardo, I have made the attached diagram showing equivalent distortions of the first twelve letters of the alphabet. The chance of such a structured pattern occurring in the same order as the stars in the sky is miniscule.
How Leonardo used the signs as templates.GIF

I looked at the diagram and have trouble understanding what you mean. But just one small thing. My immediate simple question to myself is, is the Italian alphabet the same as English? And I looked it up and apparently they don't use J, so why would Leonardo have used J? He was anticipating that English would become a world language? To be honest, I'm not even sure how they used the alphabet in Renaissance Italy, so I could be completely wrong about this, but it's just a first question.

[grapes]

I looked at the diagram and have trouble understanding what you mean. But just one small thing. My immediate simple question to myself is, is the Italian alphabet the same as English? And I looked it up and apparently they don't use J, so why would Leonardo have used J? He was anticipating that English would become a world language? To be honest, I'm not even sure how they used the alphabet in Renaissance Italy, so I could be completely wrong about this, but it's just a first question.

I thought Robert was just using the alphabet letters to show the associated distortions, rather than that Leonardo used the letters himself.

ETA: but I could be wrong

[Jens]

I thought Robert was just using the alphabet letters to show the associated distortions, rather than that Leonardo used the letters himself.

ETA: but I could be wrong

No, I see now that you're right. That's why I used as my first sentence, "I'm having trouble understanding what you mean." Clearly, I was!

[Robert Tulip]

Your comment there, that it's 12 to the 12th, seems to imply that there is a 1/12 probability of each correspondence, raised to the 12th power. Is that it?

But as near as I can see, if you allow complete freedom to distort like that, it's not 1/12, but it's near certainty that you'll be able to fit--that is, it's near certainty that you can fit any random pattern that way.

My reasoning for using the 12^12 was to indicate the extraordinarily high correspondence, starting with the hypothesis that each figure matches a zodiac constellation in order from right to left.

If you look at the first figure, on the right hand side, and ask which constellation shape it is most clearly embedded in it, the answer is Aries. So far the match to the hypothesis is a chance likelihood of 1/12

The second figure is most like Taurus. Cumulatively, the chance of this is 1/144

The third figure is most like Gemini - 1/12^3. Continuing through all twelve, we find that all twelve figures match their respective constellation, giving a chance likelihood of 1/12^12.

Here are diagrams matching each of the twelve figures in more detail to their respective constellations.

To me it is obvious that this was Leonardo's hidden intent. I am just surprised that other people cannot see it when it is so obvious.

Grapes was right that I was just using the alphabet letters to show the associated distortions, rather than that Leonardo used the letters himself. The idea is, compare the figure in the painting with the constellation, then distort the letter in the same way Leonardo distorted the star pattern. Any other ordered list would have served just as well, for example the numbers 1 to 12.

[Strange]

If you look at the first figure, on the right hand side, and ask which constellation shape it is most clearly embedded in it, the answer is Aries. So far the match to the hypothesis is a chance likelihood of 1/12

But the trouble is you are making arbitrary transformations to constellations and matching them to arbitrary parts of the picture. You need to do more than just show that you can match the constellations as you do, you need to demonstrate that they can't be mapped any other way. For example, why not use rotate Taurus 90 degrees and map it on to the thumb and forefinger in this figure, rather than the one you do use? If you used the patterns in the constellations untransformed and you always mapped them on to the posture (or the hands, or the face) then the idea might have some credibility.

I agree with grapes that the probability of a match is pretty much 1.0; I don't think it is worth the effort, but I'm absolutely certain that one could take the first 12 letters of the alphabet (or any random pattern of dots) and distort them to match each to some part of each figure.

[grapes]

I agree with grapes that the probability of a match is pretty much 1.0; I don't think it is worth the effort, but I'm absolutely certain that one could take the first 12 letters of the alphabet (or any random pattern of dots) and distort them to match each to some part of each figure.

I think what we'd have to do to convince Robert is take some arbitrary ordering of the constellations and distort them to fit the painting.

But another thing that gives us leeway, is the shape (and connecting lines) of the constellations are not standardized--or weren't, in Leonardo's time--I think.

[eburacum45]

The ancient practice of imagining pictures in the patterns of the stars ...

Before we go any further, I should state that I am very skeptical about the extent of this 'ancient practice'. Most examples I've seen look like paredoilia to me. On the other hand there are a number of ancient images that are unquestionably images of the sky, such as the Farnese Atlas
http://en.wikipedia.org/wiki/Farnese_Atlas
so obviously graphic images of the sky were well-developed and sophisticated by Classical times. Note that even for the Farnese Atlas the accuracy of the star-map is somewhat compromised.

[Robert Tulip]

Thanks. If we look at it by comparison to a password, if there are 95 keys on a keyboard including capitals, the random chance of guessing a n letter password is one in 95 to the power of n. So for a three letter password there are 857,375 different options. For a twelve letter password there are more than half a trillion trillion options.

In this example I am saying there are just twelve letters in the alphabet, and those twelve letters are the actual shapes of the star patterns that the sun passes in front of each month. It has nothing to do with the pareidolia whereby the constellation of Hercules was imagined by the ancient Greeks to somehow look like the hero Hercules. It is just the actual star patterns that Leonardo used as his template. All twelve figures in the painting prominently display the shape of the stars in the actual order that they appear in the sky. Obviously this was concealed artfully enough that it has not been noticed for 500 years, but perhaps that is just because nobody looked before.

Scientifically, a probability could be demonstrated using a matrix with the twelve figures in the painting as one axis and the twelve shapes of the ecliptic constellations as the other axis. If the rule stated that the constellation could be scaled or flipped or rotated in any way but had to retain its shape, then each of the 144 cells in the matrix could be assigned a percentage rank, from 0%, no possible match, to 100%, perfect match. My assessment is that the twelve cells diagonally down from top left would range from 70 to 90%, while almost all the rest would be less than 5%.

[caveman1917]

My reasoning for using the 12^12 was to indicate the extraordinarily high correspondence, starting with the hypothesis that each figure matches a zodiac constellation in order from right to left.

If you look at the first figure, on the right hand side, and ask which constellation shape it is most clearly embedded in it, the answer is Aries. So far the match to the hypothesis is a chance likelihood of 1/12

The second figure is most like Taurus. Cumulatively, the chance of this is 1/144

The third figure is most like Gemini - 1/12^3. Continuing through all twelve, we find that all twelve figures match their respective constellation, giving a chance likelihood of 1/12^12.

Besides the curve fitting thing, it's actually 1/12! since they aren't independent. You can't have 12 Tauri, can you?

ETA: and in your case it's even 1/11! since you're only differentiating over order, not starting point (ie 12*1/12! since you can choose where the first one is in any of 12 spots while still proclaiming the hypothesis to hold).

[astromark]

No. Leonardo da Vinci did not do this as prescribed here... This is not based on the writings and notes of the artist.
People with far to much spare time allocated odd little patterns to the stars overhead.
I blame the Greeks for starting this nonsense.. but every culture on earth has had a shot at this...
The locals around here had a fish hook as Scorpio... I see that.
The rest are nothing more than the rant of drunk wine drinking Astrologers.

[Robert Tulip]

Besides the curve fitting thing, it's actually 1/12! since they aren't independent. You can't have 12 Tauri, can you?

ETA: and in your case it's even 1/11! since you're only differentiating over order, not starting point (ie 12*1/12! since you can choose where the first one is in any of 12 spots while still proclaiming the hypothesis to hold).

No Caveman, you are not right here. The hypothesis is that the ecliptic stars are embedded from right to left. Starting from the first of the twelve figures, the question is which of the twelve constellations is it most similar to. The chance of it being most similar to Aries, as predicted by the hypothesis, is one in twelve. For the second constellation, the chance that it is most similar to Taurus is also one in twelve. They are independent, because Leonardo could just as easily have had twelve variants on the constellation shape of Aries or Leo as using the actual stars of the ecliptic. So far, combining just the first two, the chance is one in 144. There are 143 other possible combinations, just on this 'most similar' test, but the painting conforms to the expected one. Continuing across, the cumulative chance that all twelve figures are more similar to their corresponding constellation than to any of the other eleven is one in 12^12.

[Strange]

The chance of it being most similar to Aries, as predicted by the hypothesis, is one in twelve. For the second constellation, the chance that it is most similar to Taurus is also one in twelve.

I'm sorry, but applying arbitrary transformations to the shape of the constellation and then matching it to an arbitrary part of the image, the probability of a match is 1.0. The probability of matching all twelve is 1.0.

[Robert Tulip]

I'm sorry, but applying arbitrary transformations to the shape of the constellation and then matching it to an arbitrary part of the image, the probability of a match is 1.0. The probability of matching all twelve is 1.0.

I did not say 'arbitrary transformations', I said 'scaled or flipped or rotated'. This means the stars of each constellation provide the same angular relation in each corresponding figure as we see in the sky. Please have a look at this diagram that I posted earlier to show the actual relation between the figures in the painting and the observed stars in the sky, indicating just how Leonardo used the ecliptic as his template.

[Strange]

I did not say 'arbitrary transformations', I said 'scaled or flipped or rotated'.

And is there any consistency or method in the scaling, flipping or rotating? No? In that case it is arbitrary. You can match any pattern to an arbitrarily chosen part of any image in this way.

[Robert Tulip]

And is there any consistency or method in the scaling, flipping or rotating? No? In that case it is arbitrary. You can match any pattern to an arbitrarily chosen part of any image in this way.

The consistency and method is that Leonardo da Vinci took the first constellation of the ecliptic, Aries, and embedded the shape of it in the rightmost figure in the painting. He took the second constellation and embedded it in the second figure. Etc through to the twelfth figure, in order. I am not making it up. I explained each correspondence in detail here

You cannot "match any patterns" in this way. If the constellation is shaped like a Y (eg Cancer), you cannot scale or flip or rotate it to look like the Hyades and Pleiades (Taurus). And yet, the fourth figure is shaped like a Y (Cancer) while the hands of the second figure embed the Hyades and Pleiades (Taurus). Exactly the same principle applies to all twelve figures. Each figure matches most closely to its corresponding constellation. And it is not an arbitrary set of shapes, it is the shapes of the star groups which are behind the sun each year, in order.

[Strange]

You cannot "match any patterns" in this way.

Of course you can.
Seven_Samurai.jpg
I only spent about 5 minutes on that so the matches are pretty rough. But actually, I think they are nearly as good as yours (better in some cases).

And yet, the fourth figure is shaped like a Y (Cancer) while the hands of the second figure embed the Hyades and Pleiades (Taurus)

Why the figure for the 4th but the hands for the second? Can you spot a Y shape in the hands of any of the other characters? Oh, yes. So why not put Cancer there?

Each figure matches most closely to its corresponding constellation.

Do you think that if you gave someone the constellation patterns (in a random order) and asked them to match them to shapes in the picture - remember, they can match each pattern to any part of each figure - that they would put them in the right order? That would be an interesting experimental test of your idea.

And it is not an arbitrary set of shapes, it is the shapes of the star groups which are behind the sun each year, in order.

I didn't say they are arbitrary shapes. I said you perform arbitrary (linear) transforms of each one and you match them to arbitrary parts of each figure.

[Strange]

BTW, if you had some independent evidence, for example a note from Leonardo saying "Hey guys, I inserted the signs of the zodiac in this picture see if you can spot them!" then I might think you had a plausible case. But it would still be only one of many possible ways of matching them up. As it is, you have fallen into the Bible Code fallacy: look for enough matches and eventually you will find something.

The consistency and method is that Leonardo da Vinci took...

No. I meant where is your consistency and method? Do you have any rationale for the different transforms applied or why in some cases it is the head or the hands that "fit"?

[Jens]

No Caveman, you are not right here. The hypothesis is that the ecliptic stars are embedded from right to left. Starting from the first of the twelve figures, the question is which of the twelve constellations is it most similar to.

OK, I have a question then. I'm looking at the drawing you made in the first post. I'll just start from the left. I guess that's Pisces? I'm not much of an expert on the signs, I'm afraid, but just taking the shape, to me it accords most perfectly with the figure in the middle (Jesus?). The shape is almost perfect. But it also seems to go well with the fourth guy from the left, as well as the guy in red, the fourth one to the right of Jesus, as well as the second guy from the right. The third constellation from the left (is that Aries?) seems to match almost perfectly the fourth guy from the left. And for example, to me Leo looks a lot like the third guy from the right, in blue. Don't you think?

[Robert Tulip]

BTW, if you had some independent evidence, for example a note from Leonardo saying "Hey guys, I inserted the signs of the zodiac in this picture see if you can spot them!" then I might think you had a plausible case. But it would still be only one of many possible ways of matching them up. As it is, you have fallen into the Bible Code fallacy: look for enough matches and eventually you will find something. No. I meant where is your consistency and method? Do you have any rationale for the different transforms applied or why in some cases it is the head or the hands that "fit"?

There are broader cultural questions here, as to why Leonardo would have done this, and why he would have kept it a secret. It requires that we understand some of the cultural context. I have to explain this solely to demonstrate the astronomical question of how and why this empirical astronomical image of the cosmic star path of the sun could have been embedded in this painting. The intent is just to use the texts as historical documents in accordance with BAUT policy.*

On the secrecy issue, debates about science were very live in his day. Leonardo would have been well aware of the fate of his compatriot Pico Della Mirandola who it appears died as a result of promoting natural philosophy, just a decade before Leonardo painted The Last Supper. Associating the stars of the zodiac with this painting would have been immensely controversial, so his lack of discussion of it is to be expected. There are many similar discreet and prudent actions, most notably Leonardo's use of mirror writing.

As to Leonardo's motives, you have to understand there was a widely held tradition that associated traditional motifs of twelve figures with the zodiac, especially the twelve tribes of Israel and the twelve stones of the breastplate of the high priest of Israel. According to the Jewish Encyclopedia, the philosopher Philo, the historian Josephus and the theologian Thomas Aquinas all wrote that the twelve precious stones on the breast are the twelve signs of the zodiac. In the Bible, Exodus says these twelve stones symbolise the twelve tribes of Israel. Standard Biblical commentaries state that the twelve foundation stones of the holy city represent both the twelve tribes of Israel and the twelve signs of the zodiac in reverse.

Now, what is the association with this painting? Many classical works of art took inspiration from Biblical texts. One text that directly relates to The Last Supper is Luke 22:30 " you may eat and drink at my table in my kingdom and sit on thrones judging the twelve tribes of Israel."

All this would have been well known to Leonardo in his study of the motifs depicted in The Last Supper. The natural philosophy dimension, seeing the path of the ecliptic as the cosmic framework for the earth, is a further aspect, emerging from Leonardo's interest in hermetic philosophy. In his Notebooks, Leonardo included the rather cryptic comment "Hermes the Philosopher". While he was among the first modern empirical scientists and was hostile to astrology as a form of charlatanism, he nonetheless shared the interest common to learned men of his time in all the wisdom of the ancient world.

So, in considering motive, we find there is an ancient tradition among leading Judeo-Christian thinkers whereby the twelve apostles are readily associated with the twelve signs of the zodiac, through their common association with the twelve tribes of Israel. We also see that the offer by Christ to the twelve apostles to eat and drink at his table, where he says this is in order to judge the twelve tribes, is the inspiration for the image of The Last Supper. If the twelve apostles are the judges of the twelve tribes, they are also the judges of the twelve signs, in terms of the ancient association seen in Philo, Josephus and Aquinas. Given that the zodiac stars mark the empirical path of the sun through the twelve signs, it appears that Leonardo wished to conceal a message in his painting to this effect, that the symbolic Biblical story depicted overtly in the painting refers also to a natural context of empirical observation of the cosmic framework of the earth.

* cf use of historical documents