dante stella stories photographs technical guestbook

Grissom: We're not talking about sending a monkey to do a man's work. We're talking about sending a man to do a monkey's work.
Automation vs. Automatons

The human brain has been shrinking since the time of H. neanderthalensis. The apparent cause is the rise of written and pictographic language, which relieve the necessity to have so much grey matter. Some photographers claim that the use of autoexposure, autofocus, and auto-advance (or even their availability on modern cameras) are evidence of this shrinkage (or, alternatively, its causes). The catch is that the aforementioned aspects of photography often leave little room for free will - only a choice of what will be the prime mover physically.

1. Film really sets the stage for photographic determinism, and loading a roll of it - particularly a roll of 24 or 36 in a 35mm camera - commits to an initial set of variables (color or black and white? low, medium or high contrast? what kind of curve?). The ASA (now ISO) rating of the film and its characteristic curve operate as the baseline for everything else that happens. That rating is the setting at which - in the manufacturer's view - the toe and shoulder are in the right places and relationship, and the right range and linearity exists from low tones to high. It is possible to manipulate things at the developing stage, but these only result in a different fixed universe, which might - depending on the treatment - result in blocked shadows, shouldered highlights, or unreal midtones. And the expedients used - push-processing, pull-processing, or standing development - may work for some pictures on a roll but seriously compromise others.

Many photographers shoot for the normal in most scenes: lighting on the main subject and what looks to the eye like a normal range of tones from low to high (call it the "98%"). Once a photographer is locked into a film, the "normal" scene requires an exposure within a fairly narrow range to work best. Averaging meters (in-camera and handheld) get the exposure into this range - after which post-processing or optical printing calibrates that last little bit. Automatic cameras achieve the 98% of "normal" scenes fairly handily (witness that pictures in the western world are now predominantly taken with iPhones that have zero manual exposure control). Even the information that used to be printed inside film boxes could get you close enough during daylight hours. It is difficult to argue that manual controls provide any additional advantage; regardless of whether you set both variables (aperture and shutter speed) or set the more important one to you (say aperture) and let the camera choose the other (let's call this single-variable AE), you are still operating within a finite range of aperture-shutter pairings. Using a camera in manual mode does, however, create an ability to sometimes miss exposure in this normal 98% of scenes - so if your best case is getting to the same result as with an automatic, just slower, then manual cameras may even seem to be a bit inferior.

The other 2% are situations that are often induced intentionally for aesthetic purposes: shooting into the sun, shooting low- or high-key, basing the entire exposure off the darkest shadow or brightest highlight, or taking multiple spot readings to define the lighting range of a scene. These are situations that require, to varying degrees, more than simply pointing some kind of averaging meter at a subject or following the Sunny-16 rule. Situations like these absolutely require the photographer to think about the desired final result. It is far from clear, even in these situations, that shooting a manual camera (or an automatic camera in M) presents any unique advantage aside from setting a fixed exposure that never changes. Even the original single-variable AE systems, like the Konica Autoreflex T, allowed a reading to be locked, permitting the user to meter the foreground and largely ignore the backlit portions of the frame. Subsequent cameras featured exposure compensation dials that allowed a consistent exposure offset through a session (and even before that, the ISO setting could be adjusted up and down). Multipattern metering cuts into the idea of multiple independent spot readings, and exposing for the shadows (or highlights) was vastly simplified by the advent of in-camera spotmeters. Although at some gut level, automation may seem to be useless, plenty of it is well-suited to serving creative goals.

Manual-only cameras also tend to have some exposure-related features that actually degrade their usefulness for careful work:

  • Some metered manual cameras have primitive match-needle (or -diode) readouts that actually obscure what is happening. Using any Leica rangefinder, for example, gives you only indications that the exposure is one or more stops off, that it is half a stop off, or that it is on. All the time you are turning the shutter speed dial or aperture ring to make it match, you are disconnected from both numeric values. Although some people may successfully memorize the aperture ring direction and mentally keep track of where it is set, some shutter speed dials turn all the way around, and for most, this is an invitation to guess and randomly attain (or not) the correct exposure. This actually has all of the randomness of a camera running in Program mode - only there is no consistent program. Some cameras display the settings at the same time they display the match needle or diodw (like a Nikon FM2n), but the match-needle still acts only as the roughest indicator.
  • Meterless manual cameras beg the use of handheld meters, which almost universally are averaging meters (either incident or reflected). Not only does the use of these devices inject an extra step into the process, it is often also hard to tell - given the questionable ergonomics of most reflected meters - exactly what they are measuring.
  • Many manual cameras only have whole stops for apertures and shutter speeds. High precision, high thought work may want an intermediate speed or intermediate f/stop. Cameras that set these variables tend to be able to set them continuously or in one-half to one-third stops.

And there is rarely a reason to choose a manual camera over an automated one with manual override (assuming the automatic one otherwise does what you want to do - and recognizing that sometimes the manual camera is the one you happen to own).

2. Manual focus seems illusory. Although it is true that manual focus lenses are actuated by human hands, almost all modern (post 1970) manual-focus cameras utilize focusing aids: split-image rangefinders, fresnel spots, split prisms, or aerial crosshairs. All of these aids are located in the center of the viewfinder (granted, the focused subject can then be moved off-center), but more importantly, all of them operate in binary mode: they either show objects as (1) in focus or (0) out of focus. They all drive binary decisions by the photographer ("is this in focus?"). Aside from moving lens elements via twists of the hand, there is no difference between any of the systems above and a phase-detect AF reticle, a contrast-detection scheme aimed at a particular spot, or active infrared rangefinding at a defined point in the viewfinder field. Against this, active and multipattern AF also have the ability to do something a human cannot: instantly determine - and focus on - the closest object in the frame. Cameras with manual focusing modes do allow fixed zone focusing for landscapes, but most serious AF cameras also allow the AF to be switched off. And aside from the rare completely unadorned groundglass focusing screen, no design - manual or automatic - provides a realistic picture of what is in focus with the lens stopped down (don't kid yourself - that fresnel-brightened screen blacks out before it shows you the depth of field).

3. Manual winding adds no value. It is always in the photographer's interest to have an unexposed frame of film behind the shutter and ready for action. If one looks to manual winding primarily for "contemplation," the exorbitant price of film today is now the better reason to contemplate each shot. Manual winding has, for many years, satisfied an engineering consideration: power-consumption. The winding stroke(s) both moved the film and cocked a spring-driven shutter. As shutters took on more accurate electronic timing and ultimately, servo motor drive, the value of that thumb lever began to decline. By the mid-1990s, all pro 35mm cameras were motorized, and the only serious cameras left with manual wind were nostalgia-driven rangefinders, legacy designs, or medium-format cameras whose prodigious size and bulk would be compromised further by motors (or that were so slow in use that faster film availability was not a big concern). The preponderance of auto-advance cameras is not unlike the domination of Olympic shooting by semiautomatic target pistols: you need to keep your eye on the target.

None of this is to say that photography is otherwise preprogrammed; it is just to point out that performing some things "manually" is not very meaningful if it invariably reaches the same result that a robot camera would have achieved in the first place. Acceptance of technology allows you to use your diminished H. sapiens brain capacity on composition, focus selection (rather than execution), tone balance, and lighting - all of which are at the heart of a picture. And there are plenty of legitimate reasons to select a manual-only camera - but most of them have to do with unique capabilities, film format or other qualities that carry through to the final image. Anything else verges on conflating the happenstance of ancient mechanical engineering with some Zen stroke of genius.