The influence of pupil alignment on spectator address in Manet’s portraiture.

Abstract

Participants judged 94 portraits painted by Édouard Manet (70), Gustave Courbet (12) and Henri Fantin--Latour (12) for horizontal and vertical pupil misalignment and gaze ambiguity (Experiment 1) and focal point of gaze (Experiment 2). Eye movements were also measured as participants considered the extent to which sitters in the same portraits acknowledged viewers (spectators; Experiment 3). The results showed Manet portraits to be frequently painted with misaligned pupils that are associated with gaze ambiguity, especially when misaligned on the vertical axis. This ambiguity of gaze was associated with the average focal point of gaze as being judged further up and to the left of the centre for ambiguous relative to non--ambiguous portraits. These decisions in relation to portraits displaying ambiguous gaze were associated with increased eye--movements to the eye region relative to those portraits not displaying ambiguity. Finally, ratings of acknowledgement taken in Experiment 3 correlated with those of gaze ambiguity taken in Experiment 1. The results are interpreted in terms of the role of eye gaze in influencing spectatorship of portraits and, specifically, Fried's theory of the 'double relation' (Fried 1980; Fried 1996) between painting and spectator in the paintings of Manet.

The spectatorship of portraits by naïve beholders.

“…Ratings for the 94 portraits used in Donnelly et al (2017) correlate significantly with those gained in the present study (r(94) = .59, p < .001). In addition, tests using the normal approximation to the binomial distribution confirm the findings of Donnelly et al (2017) that gaze ambiguity is more common in portraits by Manet than in the combined set of Courbet and Fantin-Latour (z = 2.13, p < .05). Splitting the Courbet and Fantin-Latour set shows this contrast with Manet to remain significant for Courbet (z = 6.38, p < .001) but to be only a trend for Fantin-Latour (z = 1.70, p = .089).…”

“…Sitter gaze exerted strikingly little influence on the eye movements performed to faces or contexts during spectatorship. We hypothesized that gaze ambiguity would allow increased inspection of the context, in part because of the challenge that such gaze poses for determining the sitters focal point of attention (see Donnelly et al, 2017, Experiment 2). However no such effect was found.…”

“…In a pretest, a different set of 16 participants (four males and 12 females; M = 20.69, SD = 2.57) was presented with the 142 paintings in random order on a computer screen. They were asked to rate on a scale of 0 ( no ambiguity ) to 5 ( ambiguous ): “How ambiguous is the gaze of the primary figure in the portrait?” They received the set of standardized instructions from Donnelly et al (2017) to outline ambiguous gaze (see the Appendix). Paintings were shown one at a time so were rated separately.…”

“…When the focal point of gaze is ambiguous it is hard to classify sitter gaze as absorptive or theatrical and some artists have exploited this to create a specific effect. The portraiture of Edouard Manet is striking with respect to the frequency with which he represented gaze in an ambiguous manner by painting misaligned pupils (Donnelly et al, 2017). The effect of ambiguous gaze is to create the double-relation between spectators and sitters creating sitters that address spectators but are disengaged from them (Fried, 1980).…”

“…In the case of ambiguous gaze, where is difficult to determine the focal point of attention, sitter gaze may be a poor cue for informing inspection of the painting beyond the sitter. Here we test whether gaze that unambiguously cues attention (i.e., is focussed) to a location limits the inspection of the context of portraits relative to when gaze is a poor cue for attention (i.e., is ambiguous; Donnelly et al, 2017).…”