The Face of God

Let’s try a little test. Which of these faces looks most like God’s?

a) Milo

b) The Mona Lisa

c) Albrecht Dürer, Self-Portrait

d) Our Lady of Guadalupe

According to Joshua Conrad Jackson, Neil Hester, and Kurt Gray, in their recently published study “The faces of God in America: Revealing religious diversity across people and politics” (PLOS One June 11, 2018), most Americans would answer a) Milo.

Okay, they don’t say Milo as such. But look at the description that they do give:

What does God generally look like to American Christians? Participants saw God’s face as more masculine, Caucasian, attractive, intelligent, and loving compared to His anti-face, ts > 7.53, ps < 001 (see S1 Table for full statistics). See Fig 3. God’s face was also rated as significantly younger than the alternative composite, t = 31.83, p < .001, and as no more powerful, t = .47, p = .64, consistent with a general tendency for Americans to believe in a God who is more loving than stern. Importantly, these differences were unbiased by the characteristics of the reverse correlation base image, since we compared faces that participants selected from those they did not select. 

Masculine, Caucasian, attractive, intelligent, young, more loving than stern, less powerful than more: check, check, check, check, check, check, check.

Gosh darn it, it’s Milo to a t!

Oh, but—you will say—that can’t be right. This is science! Jackson, Hester, and Gray ran regressions and did tests and weighted for bias across all sorts of dimensions: men vs. women, liberals vs. conservatives, Caucasians vs. African Americans, young vs. old, attractive and unattractive (by self-report), Southerners vs. Northeasterners vs. Midwesterners vs. Westerners. They also used a fancy new technique called “reverse correlation,” presenting their experimental subjects with contrasting pairs of faces and asking them at each iteration to choose from the pair which image matched better the descriptions they were testing for. No way that they would end up with...Milo!

To be sure—they admitted—they asked only Christians. But, hey! They didn’t ask their 511 American Christians what Jesus looked like. They asked them what God looks like. If Christians default to thinking about Jesus, it isn’t the experimenters’ fault:

Finally, some participants may have thought of Jesus during our reverse correlation procedure, which would explain why they visualized God’s face as more loving but not more powerful than the “anti-God” face. No participants admitted to using Jesus’s face when asked to freely report any difficulty with the study, but this may have occurred outside of their awareness, or else people may not have seen it as a difficulty. To some extent, the potential overlap between God and Jesus in our measure is inevitable because many Christians believe that God and Jesus are tightly bound together (i.e., the hypostatic union). And more generally, Jesus-God confluence is an artifact of any scale that measures views of God using anthropomorphic qualities. Nevertheless, this artifact does not undermine the validity of our central results—Christians’ views of religious agents are influenced by their political orientation and egocentrism.

Staggering, isn’t it? The political orientation and egocentrism...of the experiment.

“Many Christians believe that God and Jesus are tightly bound together”? You don’t say!

Um.

Why else would anyone set up an experiment asking what the face of God looks like? ‘Cause, you know, it’s obvious that God has to look like a human being. Right? Right? Hmmm....

Let’s try another test. Which of these images looks most like God?

e) The earth

f) Ganesha

g) A crowd

h) Mark Rothko, No. 17 

I’ll let you think about it for a bit.

Over at Vice: Motherboard, Samantha Cole reported on the mind-boggling conclusion suggested by Jackson, Hester, and Gray’s comparative study of faces:

“People’s tendency to believe in a God that looks like them is consistent with an egocentric bias,” Professor Kurt Gray, the study’s senior author and a psychology professor in the College of Arts and Sciences at UNC-Chapel Hill, said in a press release. “People often project their beliefs and traits onto others, and our study shows that God’s appearance is no different—people believe in a God who not only thinks like them, but also looks like them.”

So, of course, according to the UNC-Chapel Hill study, liberals see a God who is “more African American and more loving than the conservatives’ God...reflecting their motivation for a God who encourages tolerance,” while conservatives perceive God as “more masculine, older, more powerful, and wealthier” than liberals do. Astonishingly,

though many Christians claim that God’s appearance is unknowable, our sample of believers did appear to have stable representations of God’s face that included differentiable physical features (e.g. masculinity, youthfulness, and Whiteness) and psychological characteristics (e.g. lovingness).

Believers also seem to have been convinced that God might be understandable in human terms. Go figure.

Can you tell which is the liberals’ and which the conservatives’ God? Neither can I.

I mock not because the level of theological ignorance displayed in this study is so astonishing, but rather because it is so utterly banal. The one theological justification that the paper cites for its expectation that people should look for themselves in the divine comes from the 6th-century B.C. pagan philosopher Xenophanes:

Yet if cattle or horses or lions had hands and could draw, and could sculpt like men, then the horses would draw their gods like horses, and cattle like cattle; and each they would shape bodies of gods in the likeness, each kind, of their own.

But, of course, even in Xenophanes’ day, this expectation was not a given. Think Horus with his falcon-head or Set with his composite beast-form. To expect that God should manifest himself to his creatures as a human being already presupposes an entire mythology—not to mention, cosmology— which the UNC-Chapel Hill researchers take for granted. And dismiss.

What did they expect American Christians to think God looks like? Why, a white-haired old man, of course, just like God the Creator as depicted in Michelangelo’s famous painting.

How might their experiment have been different if they knew more not just about theology, but also about the history of Christian art?

Third time’s a charm. Which of these images looks most like God?

i) The icon of Christ from Saint Catherine’s, Mount Sinai (6th century A.D.)

j) The Evangeliary of St. Andreas, Darmstadt, Hessisches Landesmuseum, AE 679, fol. 126v (11th century A.D.)

k) Saint John resting his head on the heart of Jesus, unknown German master, Museum Mayer van den Bergh, Antwerp, Belgium (ca. A.D. 1320)

l) The Rothschild Canticles, Yale University, Beinecke Rare Book and Manuscript Library, MS 404, fol. 90r (turn of the 14th century)

The correct answer to all three quizzes is, of course, m) All of the above, with the possible exception of f), depending on your understanding of hybridity.

I would be ecstatic indeed if you could tell me why.

H/t Milo for the reference to the article in Vice