Once I encountered a student defending a thesis that dealt with a topic related to mass communication. He claimed that he had conducted a “survey” of the TV audience among workers in a certain region. In reality, he had tape-recorded a dozen interviews of commuters during two train trips. Naturally the resulting transcriptions of these opinions could not constitute a survey, not only because they lacked standards of verifiability, but also because of the banality of the results. (For example, it is predictable that the majority of 12 Italians will declare that they enjoy watching a live soccer game.) Consequently, a 30-page pseudo-survey that concludes with such predictable results is a joke. It also constitutes self-deception for the student, who believes he has acquired “objective” data, while he has only superficially supported his own preconceived opinions.
A political thesis in particular risks superficiality for two reasons. First, unlike a historical or philological thesis that requires traditional methods of investigation, a thesis on a specific current social phenomenon often requires the student to invent his methodology. (For this reason, the process of writing a historical thesis may seem serene compared to that of a good political thesis.) Secondly, a political thesis risks superficiality because a large segment of “American-style” social research methodology has fetishized quantitative statistical methods, producing enormous studies that are dense with data but not useful for understanding real phenomena. Consequently, many young politicized people are skeptical of this “sociometry,” and they accuse it of simply serving the system by providing ideological cover. But people who react this way often end up doing no research at all, and their thesis becomes a sequence of flyers, appeals, or purely theoretical statements.
We can avoid this risk in various ways, including consulting “serious” works on similar topics, following the practices of an experienced group of activists, mastering proven methods of gathering and analyzing data, realizing that surveys are long and expensive and cannot be conducted in just a few weeks, etc. But since the problems presented by a historical thesis vary according to different fields, different topics, and students’ skills, it is impossible to give generic advice. I will therefore limit myself to one example. I will choose a brand-new subject on which no research has previously been done; that is of great topical interest; that has unquestionable political, ideological, and practical implications; and that many traditional professors would define as “purely journalistic”: the phenomenon of “free radio” stations.3
2.6.3 Treating a “Journalistic” Topic with Scientific Accuracy
As most Italians know, scores of these stations have appeared in large Italian cities. There are a few even in centers of a hundred thousand inhabitants, and more continue to appear across Italy. They can be political or commercial in nature. They often have legal problems, but the legislation regarding these stations is ambiguous and evolving. In the period between the genesis of this book and its publication, the situation will already have changed; as it would change during the time it would take for a student to complete this hypothetical thesis.
Therefore, I first must define the exact geographical and chronological limits of my investigation. It could be as limited as “Free Radio Stations from 1975 to 1976,” but within those limits the investigation must be thorough and complete.
If I choose to examine only those radio stations located in Milan, I must examine all the radio stations in Milan. Otherwise I risk neglecting the most significant radio station in terms of its programs, ratings, location (suburb, neighborhood, city center), and the cultural composition of its hosts. If I decide to work on a national sample of 30 radio stations, so be it. However, I must establish the selection criteria for this sample. If nationally there are in fact three commercial stations for every five political radio stations, or one extreme right-wing station for every five left-wing stations, my sample must reflect this reality. I cannot choose a sample of 30 stations in which 29 are left-wing or 29 are rightwing. If I do so, I will represent the phenomenon in proportion to my hopes and fears, instead of to the facts.
I could also decide to renounce the investigation of radio stations as they appear in reality and propose an ideal radio station, much as I tried to prove the existence of centaurs in a possible world. But in this case, the project must not only be organic and realistic (I cannot assume the existence of broadcasting equipment that does not exist, or that is inaccessible to a small private group), but it must also consider the trends of the actual phenomenon. Therefore, a preliminary investigation is indispensable, even in this case.
After I determine the limits of my investigation, I must define exactly what I mean by “free radio station,” so that the object of my investigation is publicly recognizable. When I use the term “free radio station,” do I mean only a left-wing radio station? Or a radio station built by a small group of people under semilegal circumstances? Or a radio station that is independent of the state monopoly, even if it happens to be well organized and has solely commercial purposes? Or should I consider territorial boundaries, and include only those stations located in the Republic of San Marino or Monte Carlo?
However I choose to define the term, I must clarify my criteria and explain why I exclude certain phenomena from the field of inquiry. Obviously the criteria must be defined unequivocally; if I define a free radio station as one that expresses an extreme left-wing political position, I must consider that the term is commonly used in a broader sense. In this case, I must either clarify to my readers that I challenge the common definition of the term, and defend my exclusion of the stations it refers to; or I must choose a less generic term for the radio stations I wish to examine.
At this point, I will have to describe the structure of a free radio station from an organizational, economic, and legal point of view. If full-time professionals staff some stations, and part-time volunteers staff others, I will have to build an organizational typology. I must determine whether these types share common characteristics that can serve as an abstract model of a free radio station, or whether the term covers a series of heterogeneous experiences. Here you can see how the scientific rigor of this analysis is useful also from a practical perspective; if I wanted to open a free radio station myself, I would need to understand the optimal conditions for it to function well.
To build a reliable typology, I could draw a table that compared the possible characteristics as they appear in the stations I have examined. I could present the characteristics of a given radio station vertically, and the statistical frequency of the given characteristic horizontally. Below, I provide a simplified and purely hypothetical example with only four parameters: the presence of professional staff, the music-speech ratio, the presence of commercials, and the ideological characterization. Each is applied to seven fictional radio stations.
Table 2.1
This table tells me that a nonprofessional, ideologically explicit group runs Radio Pop, that the station broadcasts more music than speech, and that it accepts commercials. It also tells me that the presence of commercials and the abundant music content are not necessarily in contrast with the station’s ideology, since we find two radio stations with similar characteristics, and only one ideological station that broadcasts more speech than music. On the other hand, the presence of commercials and abundant music characterize all nonideological stations. And so on. This table is purely hypothetical and considers only a few parameters and a few stations. Therefore it does not allow us to draw reliable statistical conclusions, and it is only a suggested starting point.
And how then do we obtain this data? We can imagine three sources: official records, managers’ statements, and listening protocols that we will establish below.
Official records: These always provide the most dependable information, but few exist for independent radio stations. I might first look for an organization’s registration documents at the local public safety authority. I might also find the organization’s constitutive act or a similar document at the local notary, although these documents may not be publicly accessible. In the future, more precise regulation may facilitate more accessible data, but for now this is the extent of what I can expect to find. However, consider that the name of the station, the broadcasting frequency, and the hours of operation are among the official data. A thesis that provided at least these three elements for each station would already be a useful contribution.
Managers’ statements: We can interview each station’s manager. Their words constitute objective data, provided that the interview transcriptions are accurate, and that we use homogeneous criteria for conducting the interviews. We must devise a single questionnaire, so that all managers respond to the questions that we deem important, and so that the refusal to answer a question becomes a matter of record. The questionnaire need not necessarily be black and white, requiring only answers of “yes” or “no.” If each station manager releases a statement of intent, these statements together could constitute a useful document. Let us clarify the notion
