Searching for the Crucifixion (a google experiment)

As is well known, a large percentage of medieval and early modern art was made for the Church, and therefore it’s always tempting for me to do a topical post whenever one of the major Church holidays comes around. So, seeing as today is Good Friday, it’s Crucifixion time chez L’Historien Errant!

Now, with all probability the Crucifixion is the most frequently depicted subject in Christian art, rivaled only by the Virgin and Child. Indeed many late medieval painters’ guilds regulations require apprentices to deliver a Crucifixion and/or a Virgin and Child as their masterpiece(s). That means I had plenty of material to choose from – one might even say: too much material to choose from – and for a while I really couldn’t decide just which Crucifixion(s) I was going to blog about. So in the end I decided to let Google do the work for me: I was going to do a Google image search and pick whichever depiction of the Crucifixion came up as the first search result (though ignoring film stills from that Mel Gibson movie, photos of Good Friday processions in Latin America, and suchlike). There was one more thing for me to decide, however – was I going to conduct the search in English (the blog’s language) or in German (my native language)? And that was the point at which inspiration struck me and made me think up the following experiment: I was going to do an image search for Crucifixion in all the major (Western) languages and compare the results.

I started with English, and was in for my first surprise. The highest ranking result for “crucifixion” turned out to be this

Marco Palmezzano, Crucifixion, ca. 1480/90, Galleria degli Uffizi, Florence (Image © Wikimedia Commons)

… an admittedly fine but nonetheless rather obscure painting by a rather obscure artists, Marco Palmezzano (1460 – 1539) who spent almost his entire career in the small Italian town of Forlì. I must confess I’d never heard of him before…

Sticking with the same search term but changing Google’s default language settings, I next searched for French “crucifixion”. This time a more prominent painting by a much more prominent artist was ranked first, a panel Pietro Perugino (c. 1445 – 1523) painted for the church of San Domenico in San Gigmignano:

Pietro Perugino, Crucifixion, c. 1482, National Gallery of Art, Washington (Image © Wikimedia Commons)

Next was Spanish “crucifixión”, and perhaps unsurprisingly the top result was a painting which is now kept in the Prado:

Juan de Flandes, Crucifixion, 1509, Museo del Prado, Madrid (Image © Wikimedia Commons)

As his name suggests, the painter Juan de Flandes (c. 1460 – 1519) was a native of Flanders, and it is commonly assumed that he received his training as an artist in Ghent. He first appears in written sources, however, only in 1496 when he was already active in Spain, more precisely at the court of Isabella I of Castile. He spent the rest of his career on the Iberian peninsula and – together with Pedro Berruguete – is considered the most important painter of Renaissance Spain.

Speaking of the Iberian peninsula, here’s what Portuguese “crucificação” brought up:

Gaudenzio Ferrari, Crucifixion, 1513, S. Maria delle Grazie, Varallo Sesia (Image © Wikimedia Commons)

This great fresco by Gaudenzio Ferrari (c. 1471 – 1546) has long been a favourite of mine. I mean, considering that Ferrari was working in early 16th century Italy and in many respects was influenced by the Milanese works of Leonardo da Vinci, there is something incredibly medieval in the character of his paintings, so naturally my medievalist self is going to like them. That said, I guess a more objective take on Ferrari would be to view him as a rather provincial figure who overall probably isn’t any more prominent than Marco Palmezzano.*

What a contrast then to google the Italian “crocefissione”! Apparently, the Italians know what is expected of them so the top result in this search was by one of the most prominent painters of all time, Raphael (1483 – 1520) – and I believe at least one of my readers will be delighted by this:

Raphael, Crucifixion, c. 1503, National Gallery, London (Image © Wikimedia Commons)

However, in Italian there is also a second (actually more common) spelling of “crocefissione”, i.e. “crocifissione”. Ok, that’s not really much of a difference, but it made quite a difference in the search results. But again, what came up in the first place was a work by one of Italy’s most famous Renaissance masters, Masaccio (1401 – 1428):

Masaccio, Crucifixion, 1426, Museo di Capodimonte, Naples (Image © Wikimedia Commons)

But things got a bit weird again when I entered the German “Kreuzigung” as my search term. I would have sworn that the highest ranking result would have had to be one of Matthias Grünewald’s famous Crucifixions, most likely the one from the Isenheim Altarpiece. But no, what I got instead was a panel from Johann Koerbecke’s Marienfeld Altarpiece:

Johann Koerbecke, Crucifixion, 1457, Staatliche Museen Preußischer Kulturbesitz, Berlin (Image © Wikimedia Commons)

Koerbecke was active from c. 1450 to 1490 in Westphalia and was heavily influenced by that region’s leading masters of the previous generation, i.e. Conrad von Soest and Stephan Lochner. And while I personally rather like this painting, it is just as obscure, if not even more obscure than Palmezzano’s, and you really have to have a thorough knowledge of 15th century German art to have come across this particular Crucifixion – or indeed any work by Koerbecke – previously.

All in all, I guess this little Google experiment has once again proved the well-known fact that the highest ranking results in a Google search aren’t necessarily the most relevant ones. One important factor in this is of course Wikipedia which for a number of reasons always makes it to the top of the list in any search. And indeed, the surprisingly high position of Palmezzano’s Crucifixion in my image search is most likely explained by the fact that this painting actually is the first illustration you see on the entry dedicated to the Crucifixion in the English version of Wikipedia. Similarly, the first image you see when you check out Crucifixion in the French Wikipedia is the panel by Perugino shown above. However, no such direct connection to Wikipedia seems to exist with the other paintings discussed in this post, so it remains somewhat mysterious just why Google thinks they are so important (especially in the case of the Koerbecke piece).

The most interesting result for me was, however, that all the works of art that came in first in my multi-lingual search date to the period of c. 1420 – 1520. This corresponds more or less exactly to the era from the beginning of Renaissance painting in the 1420s to the end of the High Renaissance around 1520 (conspicuously marked by the deaths of Leonardo and Raphael in 1519 and 1520 respectively). This was of course a very decisive period for European art, and for the classicist tastes of the 19th and early 20th centuries this was the period when art had reached its highest level of beauty and perfection. Everything that had gone on during the Middle Ages was considered an immature, preliminary stage, everything that followed was viewed as decay. I find it quite amazing how much the results of my search still seem to reflect to 19th century’s preference for the Renaissance.

That said, there is one more unknown we have to take into the equation, namely the fact that Google auto-customizes all search results. This means that based on all your previous searches and online activities Google makes up some kind of idea of what your interests and your preferences are, and lists search results accordingly. Now, as you well know, 15th and early 16th century art is something that does indeed interest me very much, so there is the possibility that my own previous searches simply made Google rank finds from that period higher than others. But that still doesn’t explain why, for example, Koerbecke got listed before Grünewald – after all, I have recently blogged about Grünewald and even written a Grünewald-related article, so in the recent past his name appeared in my google searches quite a lot and I checked out quite a few websites and online resources featuring information about the artist and his work. If anything, therefore, my own personal search history should have given Grünewald even more prominence than he has to begin with…

But anyway, you know what would be cool? If perhaps you could spare a few minutes of your time and repeat my little experiment for yourself. I’d be really interested to see if your searches bring up the same, similar or completely different results than my own.


* As you may remember, though, I do have a soft spot for artists that are generally considered provincial and/or second-rate (see here and here). On that same note, David Packwood of Art History Today recently posted a highly fascinating account of Pordenone’s Provincialism. If you haven’t read it already, I’d definitely recommend you to do so – there’s even a rather spectacular depiction of the Crucifixion in it…

This entry was posted in Uncategorized and tagged , , . Bookmark the permalink.

10 Responses to Searching for the Crucifixion (a google experiment)

  1. Hi, what an inspiring idea… you were quite right. Googling from a Danish outpost, I found a somewhat different selection. You may view it at my interest account:

    • Hi, and thanks for joining the experiment, Karen.
      I’m glad to see that Grünewald’s Isenheim Crucifixion holds the number one spot at least in the Danish search, because that really is the painting I would have bet on…

  2. Hello HE,
    I’m travelling until 4/16 but I reserve the right to comment on this post when I have the opportunity. In the meantime see this:

    Bob Consoli

    • Thanks for the comment, Bob, and also for the link – your open letter to Google is quite thought-provoking and, I’m afraid, every bit as true and relevant today as when you wrote it.

      I’d be most interested of course to read your thoughts if (and whenever) you find an opportunity to comment again. Until then, enjoy your travels!

      • I didn’t mean to be so peremptory. What I really meant to say is ‘I really want to comment on this article!’ The plain fact of the matter is that Google Search just isn’t very good; the reliance on Wikipedia demonstrates that. More later,

  3. Shtina says:

    What a great experiment! I was expecting to see Grünewald ‘s Isenheim altarpiece but it’s the third English (USA) image that comes up.

    Since I saw Caravaggio’s The Crucifixion of Saint Peter within the first ten images, I searched “The Crucifixion” and got the Bronzino from Wikipedia as the first image. Interestingly, the first image on the Wikipedia entry “Crucifixion of Jesus” is a Vouet painting that actually shows up second in the Google image search. Kind of interesting how google clearly prioritizes Wikipedia but then non-intuitively rank the Wikipedia images.

    • Thanks for joining in the experiment :-)

      In the results I got, Grünewald’s Isenheim Crucifixion came up second or third, or at least among the top twenty results in most of the searches, too, but it’s quite curious that it didn’t make it to number one in any language…

  4. What a fun post and a neat idea! I don’t think that I would have guessed that Ferrari would come up with “crucificação” in Portuguese. When I did the word on my end in a Google Search, “crucificação” brought up Caravaggio’s “Crucifixion of Saint Peter” first (similar to what Shtina wrote).

    Did most of these images pop up in Google Image via Wikipedia? Caravaggio’s “Crucifixion of Saint Peter” is the first image that is listed under the Wikipedia entry for “crucificação.”

    • Thanks for the comment, Monica!

      The Portuguese search is really curious: I spent Easter at my parents’ house and took the chance to repeat my experiment on their computer – all the results were the same except for the Portuguese version which brought up Antonello da Messina’s Antwerp Crucifixion first (though the Ferrari fresco still came in second place).

      And yes, most of the images in Google Image came up via Wikipedia, but not all of them. For instance, the Masaccio painting came from a site called artinvest2000 – even though it is featured in the French Wikipedia entry “Crucifixion“.

  5. It is generally acknowledged, that if you want an article to pop up at the front page of a google search, you have to link to it at wikipedia (which is of course tricky because of their strict rules) or post it at a social website (Facebook, google+, trip advisor etc). And then you have to be sure to keep your site lively and “newsy”… the more articles, the higher up. I disagree with those who think this makes google problematic; what is problematic is the fact, that “they” have organized the algorithm according to what they think I am interested in. It is complicated for me to bypass their idea, that I primarily want Danish news (which I don’t)

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s