Roskill, Carrier (1983): Truth and Falsehood in Visual Images:

An im­age can­not be true or false in the sense that the propo­si­tions of sci­ence or math­e­mat­ics are true or false

Yet we do think about im­ages in terms of truth and false­hood.

Shaun Tan:

Drawing a good pic­ture is like tel­lling a re­ally good lie — the key is in the ince­den­tal de­tail

The way you cre­ate an au­then­tic-lolok­ing im­age (deception) is pay­ing close at­ten­tion to de­tail and mak­ing it look ince­den­tal.


bristol a-z Map of Bristol fea­tur­ing the fic­ti­tious Lye Close. Cabinet Magazine

Maybe the most re­li­able im­age is a map. First, maps are pub­lished and copy­righted doc­u­ments (giving them a sense of au­thor­ity). We ex­pect a de­gree of ac­cu­racy de­pend­ing on the ( kind and scale of map).

The cat­e­gories de­fended:

What would it be like to have a 1:1 scale map? This would be a map that’s so de­tailed it would need to con­tain an im­age of it­self. We now get this some­times in google Streetview (with the pho­tog­ra­phy car showin up in re­flec­tions etc.)

Also Jorge Luis Borges (1946): On ex­ac­ti­tude in Science

So: a true im­age might have a re­la­tion­ship to re­al­ity (but not in every de­tail). Maps are mod­els of re­al­ity.

Semiotics three kinds of im­ages:

  1. Index: Fingerprint
  2. Icon
  3. Symbol: Written lan­guage (in or­der to mean some­thing it does­n’t have to look like the things it’s rep­re­sent­ing)

An im­age of a chim­pansee fin­ger­print. For this im­age to ex­ist, the chim­panzee has to be pre­sent. It’s made by the thing it’s rep­re­sent­ing. (if you re-drew it, it would be­come an icon). Fingerprint as an ev­i­den­tiary im­age. We also con­flate the fin­ger­print with iden­tity - if you’re fin­ger­print is there, you were there.

Dauguerre 1838 Commons

Photographs are a dif­fi­cult case. It’s an icon, but also ev­i­den­tiary. It’s in­dex­i­cal be­cause of the ac­tion of light on a sil­ver plate. An in­dex­i­cal record of how light was bounc­ing of things. Without a cap­tion, we could­n’t be sure this was ac­tu­ally show­ing the boule­vard de tem­ple. The in­dex­i­cal re­la­tion­ship is re­ally the only re­li­able truth­ful­ness im­ages have.

Gunnig (2004): What’s the Point of an Index? or, Faking Photographs

As men­tioned last week the it’s phe­nom­ena”: We don’t look at he ma­te­r­ial re­al­ity of pho­tographs, but look through them at what they’re show­ing. Photographs were be­ing ma­nip­u­lated ba­si­clly from the moe­ment they were in­vented.

Susan Sontag (1973): On Photography

Would we pre­fer a shit pho­to­graph of shake­speare to a great paint­ing? Yes, every time. There’s this idea that a pho­to­graph can bring you into a place ina way that no other im­age can.

The first ever digital image Walden Kirsch by Russel Kirsch (1957). NIST

In the case of a dig­i­tal pho­to­graph, light is caus­ing a nu­mer­i­cal re­sponse. Reading it is es­sen­tially like read­ing a ther­mome­ter. It’s not in­her­ently less in­dex­i­cal than a film cam­era.

First image from mariner 4 A real-time data trans­la­tor” ma­chine con­verted Mariner 4 dig­i­tal im­age data into num­bers printed on strips of pa­per. Too anx­ious to wait for the of­fi­cial processed im­age, em­ploy­ees from the Voyager Telecommunications Section at­tached these strips side by side to a dis­play panel and hand col­ored the num­bers like a paint-by-num­bers pic­ture. NASA

This im­age comes from the thing as in­te­geres. The sci­en­tists would­n’t wait un­til the im­age was ren­dered, so they as­signed each num­ber a crayon colour and drew out the im­age.

Bomb cab at first 1/100000000th second Harold edgerton Harold Edgerton: Atomic Bomb Explosion atop a Steel Tower in Eniwetok. Source

We have this as­sump­tion that a photo rep­re­sents a mo­ment in time (we don’t re­ally ques­tion how long that mo­ment is). The above im­age is from a spe­cially built cam­era of an atom bomb ex­plo­sion (1/1m sec­ond shut­ter speed). It’s an ac­tion tha’s much faster than the eye could ever see.

Atomic test explosion edgerton weird peach Harold Edgerton (1946-52): Atomic Bomb Explosion. The Metropolitan Museum of Art

These im­ages cause vi­sual des­per­a­tion - you hold onto eh trees and scaf­fold­ing to ground your­self. THese are im­ages we can­not ver­ify us­ing our own ex­pe­ri­enc.e

the hubble deep field Hubble Deep Field (full mo­saic) re­leased by NASA on January 15, 1996. Commons

This is a 100 hour ex­po­sure. In a dark patch of sky, you sud­denly see these thou­sands of galax­ies.

Changbom park and juhan kim deep field image Figure 8 from Diffuse Dark and Bright Objects in the Hubble Deep Field

These deep field im­ages are en­tirely im­pos­si­ble for hu­mans to look at — Park and Kim (1997)they had to write soft­ware to make sense of them.

Six Stories from the End of Representation

Different modes of con­trast in a mi­cro­scope. In mi­croscopy you have to in­crease the con­trast to see what’s oing on - none of these im­ages are re­ally true

When the first peo­ple were look­ing through mi­cro­scopes, they saw alll kinds of stuff that was­n’t there be­cause they ahd no ref­er­ence points in their re­al­ity.

Albrecht von Haller: Uterus and Vagina

This is too de­tailed to be use­ful. All these pat­terns of tis­sue that we can’t re­alte to - the im­age stop sof­fer­ing us truth.

Stereoscopic pho­tographs of a pelvis by Cunningham

In or­der to be truth­ful, things need to be left out. Often in med­ical il­lus­tra­tion, we es­sen­tiall draw maps (geometry) of anatomy. Bodyworlds is more like a 1:1 map.

Agatha christie map at the be­gin­ning of a who­dun­nit.

Next: Court il­lus­tra­tion These are one of the few in­stances were draw­ings be­come ev­i­den­tial

Julia Quenzler:

They were us­ing pho­tographs be­fore I came on the scene twenty years ago and draw­ings proveed to be far more re­veal­ing

The only thing she could be talk­ing about ehere is sub­jec­tive (from be­ing there as a wit­ness)

kuglars peo­ple Oliver Kugler

Reportage il­lus­tra­tors of­ten talk about their im­ages in op­po­si­tion to pho­tog­ra­phy (because they ex­ist in the same space). THere’s a myth of re­portage il­lu­s­tra­tui - it looks line and wash be­cause they’re drawn fast, but con­versely we start to think im­ages that look like this mean some­one was ac­tu­ally there. (naturalised myth).

Jill Gibbon on her web­site adds a fake sketch­book be­hind her draw­ings to give au­then­tic­ity to the im­ages. It does some of the work of de­scrib­ing the process.

The ef­fec­tive­ness of black­mail pho­tographs de­pend en­tirely on the vic­times con­ces­sion that they are un­de­ni­able. (blackmail draw­ings how­ever, are eas­ily de­ni­able and would not work)

Gombrich: Art and Illusion

Henri de Toulouse-Lautrec (1893): Miss Loïe Fuller

This doesn;t show the body or records the dance, but it shows some kind of bg­ger, sub­jec­tive truth.

bonny prince char­lie / prince henry.

Captions dic­tate the way we read im­ages - the iamge has­n’t changes, but our read­ing of it has. When his­to­ri­ans dis­agree about im­ages, our idea of truth be­comes shaky


Trompe l’oeil paint­ing

Reverse side of a painting 1670. Cornelius Norbertus Gijsbrechts (1670): The Reverse of a Framed Painting. Statens Museum for Kunst

Gombrich: Art and Illusion

These aren’t a re­spected genre of paint­ing — they’re more like a magic trick.

Ames chair demon­stra­tions in per­cep­tion

From one an­gle the de­cep­tion look real, from any other an­gle they reak down.


As wittgen­stein ar­gued for wrod that the mean­ing is in the use - so for each pho­to­graph

Roger fenton the valley of the shadow of death Roger Fenton (1855): Valley of the shadow of death. Commons

Sontags ac­cu­sa­tion is that fen­ton moved the can­non balls, nyt re­porter tries to prove her wrong (and does­n’t prove any­thing). THe only thing these im­ages tell us is that fen­ton was thre and took these pho­tos. When we for­get about Fenton, the truth in these im­ages is even more dimnished.

Beyus on a forced land­ing

Had it not been for the tar­tars I would not be alive to­day … Yet it was they who dis­cov­ered me in the snow af­ter the crash, when the German search par­ties had given up.

These two im­ages ap­pear in two dif­fer­ent mono­graphs that both tell the above story (which is a lie)

Operation Mincemeat

Ian Fleming and John Godfrey

Mincemeat Major Martin ID Card. Commons

Britain and the al­lies tried to con­vince ger­many that were go­ing to in­vade sycile They gave a dead lon­don per­son all kinds of doc­u­ments - let­ters, naval iden­tity card, a suit­case con­tain­ing made up plans to in­vade ar­dinia and dumped it (using a sub) on the coast of spain.

Also the con­text in which we find im­ages change the way we read them.

[Hurricane sandy day af­ter to­mor­row] [In­stal­la­tion by su­per­flex]

The tweet cap­tions di­rect our un­der­stand­ing of these im­ages

Death in the air 1933 These were cre­ated us­ing props by a bbc em­ployee

Atlas Group (1967) by Walid Raad

He made up this group of ar­tis­sts work­ing in the lebanese civil war (all ofwhom were him) These use of a de­cep­tion to talk about some­thing that was­n’t be­ing talked about

JJ Abrams: S. (2013)

A novel with two char­ac­ters com­ment­ing on it. All these tip ins.

This cre­ates a nat­u­ralised nar­ra­tive sit­u­a­tion (suddenly we have a cred­i­ble rea­son for these im­ages to ex­ists) in lit­er­a­ture these are di­aries

Leanna Shapton: Important ar­ti­facts A love story in the form of an auc­tion cat­a­logue

Visual coven­tions used by de­cep­tions:

Lye close in bris­tol does­n’t ex­ist — they’re a copy­right trap to catch peo­ple copy­ing the map. Google maps also has these

F for Fake