I post this because Furstenau welcomed responses to the question “Is quantification the best way of getting at what is most significant in style?”
I have to assume what he implied by this question was not whether cinemetrics can or cannot be a “best way,” but instead questioned the exact usefulness of the technology.
Personally I do not have a great amount of experience with cinemetrics, and have indeed never used it myself. My first exposure to it was at last year’s annual FSAC colloquium where one of the panels consisted of three presenters who all used the technology. As examples of the method (if it can indeed be called a method) they were not wholly convincing of its value. But neither were they completely unconvincing, and I would be lying to say that I did not see tremendous potential in such a method.
I find it moot to wholly dismiss any attempt at a methodology, especially one that has not yet had much chance of proving its usefulness.
The problem for me with Furstenau’s vilification of the method is that he appeared to focus on cinemetrics as a stylistic component of the studied film itself, as though the graphical representation of whatever was being analyzed was the end result.
Granted, Furstenau did supply, as did Tom Gunning, a couple examples of the ludicrously frivolous findings that result from some such studies. And I do agree that an exploration of style should often, if not always, begin with the filmic object itself.
However, the value of the method is not in the graphical representation itself, nor – and I believe Furstenau would agree – is it worth constructing the graph only to find what one was looking for after the fact. In other words, why would one bother spending the time to compile a graph simply to start with the graph as the object of study: this is film studies, and it is infinitely more satisfying – especially in terms of style – to begin with the film itself as your object of interest.
But the tool is undoubtedly useful in terms of analyzing formal stylistics. There are numerous cinema books that would benefit from a condensed graphical representation of quantified analyses of particular films. I cannot express in words the dissatisfaction and anger I have experienced from the amount of quantitative formal analyses I have read that have resulted in no summary conclusion of why that analysis mattered. Cinemetrics may be the contemporary poster-child for the skepticism attached to empirical analysis, but it is by no means the first method to produce unsatisfactory results.
The ability to measure and identify trends in terms of production periods or genres of cinema that cinemetrics supplies is a fruitful pursuit.
This is all said in acknowledgement of the fact that cinemetrics is not an end in itself, but is a useful and fairly concise tool in coming to other conclusions inspired primarily by some stylistic element of the film. I say fairly concise because, as Martin Lefebvre has pointed out, there remains the error of the individual using the tool: if the process depends on the analyst’s ability to catalogue, we must question the legitimacy of their ability.
There are two further points Furstenau made that I’ll respond to. First of all, he stated that style is a contingent phenomenon that surfaces at the engagement between the viewer and the screen. I would simply like to point out that the careful study of a cinemetric graph might also provide the contingency required for the analysis of style. One may simply start with the graph and then move to the film. As mentioned above, I agree that this would not be a preferred method of engagement and starting with the filmic object seems more ideal. However, if contingency is the required phenomenological marker of stylistic characteristics then it is not convincing that one must start with the filmic object.
Secondly, Furstenau stated that cinemetrics provides a minimal amount of information. But cinemetrics is not a conclusion of an analysis. The analysis comes after the graphs have been established: the information is derived from those graphs in the process of the larger analysis. Regardless, this is not the only method of film that provides a minimal amount of information. It is what is done with that information that is the value of these analyses.
Ultimately, distinctions must be drawn between the project of cinemetrics and the use of cinemetrics.