|Paper given at the Nineteenth Goddard Memorial Symposium of the American Astronautical Society, Pentagon City, Virginia, March 26-27, 1981.|
|Subsequently published in: Science Fiction and Space Futures, edited by Eugene M. Emme. San Diego: American Astronautical Society, 1982, pages 121-135.|
|Beginning||Fig.1: Attitudes to Technology|
|Ideologies of SF||Fig.2: SF Ideologies|
|Impact on Students||Fig.3: Science-Technology|
|Notes||Fig.4: Attitudes of Factions|
|Fig.5: SF and Space Program|
|Fig.6: Attitudes of Students|
|Fig.7: Technology and SF|
Hugo Gernsback, founder of the first true science fiction magazines more than fifty years ago, felt that this new form of literature should serve a revolutionary purpose. He wanted "SF" to educate readers in scientific Principles, to inspire young people to enter technical professions, to sketch future inventions which readers could then perfect, and to generate enthusiasm for science and technology among the general public. There is evidence that all these effects did follow in the early days; for example, the great social movement which produced modern space rocketry was inspired by certain science fiction classics. But today there is real question whether the literature still promotes scientific and technological values, or whether it now speaks in many contradictory voices, some adamantly opposed to the ideology of its founders.
In the 1960s, a New-Wave movement in science fiction promoted avant-garde literary experimentation, criticism of technology, and an interest in the social rather than in the physical sciences. 'Chile few writers marched long under its banners, a large proportion of those active today are in some degree fellow travelers with the New Wave. In addition, there is a large and popular group of Fantasy authors who have found a secure place with the science fiction field, mainly producing what is called Sword-and-Sorcery. Ignoring science, these writers spin stories around magic and swordplay, implicitly urging technological primitivism and a reversion to our barbarian ancestors. Furthermore, whatever the ideological thrust of SF literature, there is some question whether it gets communicated successfully to the mass public, because popular visual media present only the most distorted and attenuated shadow of science fiction.
This article will analyze the ideological impact of science fiction, using data from large questionnaire surveys. The mode of analysis is standard correlational opinion research. The first of the two main studies was carried out at the 1978 World Science Fiction Convention in Phoenix, Arizona, where 595 dedicated members of the science fiction subculture were willing to complete a lengthy questionnaire. The second survey was administered in 1979 to 1,439 students at the University of Washington, leading research and teaching institution in the Pacific Northwest.
One set of six questions tapping respondents' attitudes toward technology was included in both of these surveys. The items were drawn from a study carried out in 1970 in the Boston area by Irene Taviss. She used a long list of statements, expressing attitudes both for and against technological progress, and published results of a factor analysis that permitted me to select three items which well represented anti-technology sentiments. I also used one general pro-technology item and two others on specific topics of interest: spaceflight and atomic power. Together, the six are a proven attitude measure. Figure 1 shows the percent of Taviss' Boston area respondents expressing agreement with each statement, comparable figures for the 595 SF fans, and percentages also from a spaceflight survey of 225 Seattle area voters.
|201 Boston Area Residents||225 Seattle Voters||595 Science Fiction Fans|
|1. Machines have thrown too many people out of work.||55.2%||21.3%||6.1%|
|2. It would be nice if we would stop building so many factories and go back to nature||35.8%||16.9%||15.8%|
|3. Technology has made life too complicated.||32.3%||20.9%||10.4%|
|4. Technology does more good than harm.||76.1%||65.3%||67.4%|
|5. The potential dangers of nuclear energy are outweighed by its potential benefits.||49.8%||54.7%||56.3%|
|6. In the long run, discoveries made in our space program will have a big payoff for the average person.||38.8%||48.9%||88.7%|
Many things separate the three groups. The Seattle voter survey focused on attitudes toward spaceflight, and Seattle is an aerospace town, so we might expect more pro-technology attitudes in this group. Several standard variables are known to favor pro-space attitudes, including being young, educated, middle-class, and male -- characteristics which the SF fans tend to have. Consequently, we cannot be sure about the sources of the differences between the groups in Figure 1, but this table will be a good starting point for our deliberations.
The respondents from the science fiction convention are less likely to agree with the three anti-technology statements, although there is no significant difference with Seattle voters on abandoning factories and going back to nature. The first two pro-technology items do not show much difference, however, and the SF fans are the middle group in percent who feel technology does more good than harm. But on the item about spaceflight, the science fiction subculture really stands out. Nearly twice as many fans, a huge majority, are optimistic about the long-range benefits of the space program. We cannot rely much on a single table without elaborate statistical controls, but three conclusions suggest themselves to us. First, SF fans are somewhat less likely to express anti-technology sentiments. Second, their pro-technology sentiments show a mixed pattern, not necessarily differing from those of the general public. Third, there is unusual and overwhelming support for the space program among SF fans.
To get a clear picture of contemporary science fiction, and the competing ideologies which dwell within, we shall now look at the main results of the Phoenix convention survey. Most questions were simple preference items. Respondents were given lists of 140 authors' names and 62 types of literature and asked how much they personally liked each one. They gave their answers on a seven-point scale from "0" (do not like) to "6" (like very much). There were five different editions of the questionnaire, presenting the items in five different random orders, to guard against spurious results which might have been caused by the placement of the items.
The tool of analysis was statistical correlation. If fans like author X, do they also tend to like author Y? The coefficients reported here are either Pearson's r or Kendall's tau, depending on the nature of the data, with tau generally being smaller than r simply because of the way it is calculated. These coefficients range from -1.00 to +1.00, and a coefficient significantly greater than zero indicates a positive association between two variables. If there is a significant correlation between X and Y. we can conclude they have something in common. For example, the fact that preference for the author Harlan Ellison correlates very strongly (r = .52) with preference for "New-Wave science fiction" reflects the fact that he is a prominent author of this type.
Figure 2 shows correlations between selected descriptive phrases and three main kinds of literature: Hard-Science SF, New-Wave SF and Sword-and-Sorcery. For this analysis, I wanted to use only the most honest and most knowledgeable respondents, rather than all 595. To catch incompetent or frivolous responders, I had included the names of two fake authors among the list of 140. Persons who claimed to have an opinion of either fake author, and those who failed to express opinions about at least 50 real authors, were removed from the analysis, leaving 409 good respondents.
|Hard-Science SF||New-Wave SF||Sword and Sorcery|
|Fiction based on the physical sciences||0.66*||0.05||0.04|
|Fiction based on the social sciences||0.02||0.40*||-0.08|
|Stories about magic||-0.05||-0.01||0.57*|
|Stories about new technology||0.51*||0.07||0.02|
|Stories which take current knowledge from one of the sciences and logically extrapolate what might be the next steps taken in that science||0.47*||0.03||0.05|
|Avant-garde fiction which experiments with new styles||-0.05||0.65*||0.04|
|Fiction which deeply probes personal relationships and feelings||-0.20*||0.37*||-0.03|
|Stories about barbarians||0.06||0.02||0.66*|
The traditional variety of science fiction, the kind favored by Gernsback and by authors like Robert Heinlein and Isaac Asimov, is often called "Hard-Science science fiction." Knowledgeable fans generally liked this type, giving it an average rating of 4.62 on the "0" to "6" scale, compared with 3.81 for Sword-and-Sorcery and 3.30 for New-Wave. The first three rows of coefficients in Figure 2 show correlations between these three types and fiction based on three different ways of confronting reality: physical science, social science, and magic. The extremely high coefficient (r = .66) linking Hard-Science with "fiction based on the physical sciences" essentially defines this type. Although in principle the coefficients have a ceiling of 1.00, the "noise" in survey responses prevents us from approaching this level, and .66 must be considered about as strong a relationship as we would ever expect to find. Therefore, one might justly define Hard-Science SF as fiction based on the physical sciences.
There is a smaller, but very respectable correlation (r = .40) between New-Wave and "fiction based on the social sciences." This means that New-Wave is connected with popular conceptions of the social sciences and nearly is defined by this phrase. The correlation between Sword-andSorcery and "stories about magic" (r = .57) approaches the level of definition. All three of these are very strong associations, in comparison with what we usually get in survey research. We must not take them as a standard of "good correlation," because they represent mental links on a single level of abstraction -- essentially definitions of terms. When we correlate across levels of abstraction, we must be content with lower correlations.
The next two rows in Figure 2 show that Hard-Science, but not the other two types, can also be described, to put it simply, as stories about new technology and new science. Next, we see a very powerful association (r = .65) between New-Wave and "avant-garde literature which experiments with new styles." Here we find the best definition of New-Wave. Although in part it is concerned with the topics of the social sciences, mainly it represents literary experimentation. We also see a strong correlation with "fiction which deeply probes personal relationships and feelings," a humanist literary value which actually shows a small negative correlation with Hard-Science. This means that while New-Wave fans like such fiction, Hard-Science fans show a weak tendency to rate it lower.
Finally, we see powerful, coincidentally identical correlations (r = .66) linking Sword-and-Sorcery with "stories about barbarians" and with the general term "fantasy." Although sold side-by-side with true science fiction, even included in the same SF magazines, Sword-andSorcery concerns an imaginary barbarian past rather than the future of science and technology. While Hard-Science praises science and technology, and New-Wave is more interested in literary, social, and psychological issues, Sword-and-Sorcery seeks to escape this world altogether.
Two of the items highly correlated with Hard-Science, new technology and new science, express what Gernsback and thousands of readers have always assumed science fiction was all about. We can evaluate the extent to which these topics still define the field by looking at correlations between these two items and 122 of the best-known authors in the field. A table of these coefficients would fill much of this article, so I present the data by means of a graph, in Figure 3. Each dot represents an author. The coordinates are the author's correlations with "stories about new technology" (new technology) and with "stories which take current knowledge from one of the sciences and logically extrapolate what might be the next stems taken in that science" (new science).
The authors at the upper right are all members of the Hard- Science faction, including Gernsback himself and such popular writers as Isaac Asimov and Arthur C. Clarke. The group at the lower left includes several New-Wave writers, including Harlan Ellison, Raccoona Sheldon, Joanna Russ and Ursula K. LeGuin. Closely allied with the New Wave are "mainstream" writers Ray Bradbury, Kurt Vonnegut and Stanislaw Lem. Sword-and-Sorcery is represented by Marion Zimmer Bradley, Michael Moorcock and J. R. R. Tolkien, with other writers of similar fantasy clustered around them. Detailed analysis of all the authors represented in this chart must wait for my future book about the ideologies of science fiction. But a glance at the chart shows that authors differ greatly in the extent to which their fiction is tied to an interest in science and technology.
Although the convention survey was primarily designed to map science fiction and delineate the literary ideologies within it, and space limited the number of other questions which could be included, we do have data for a preliminary look at the different propagandistic effects of the three competing types of literature. Figure 4 shows how each of the three types and each of the three ways of confronting reality correlates with three variables which plausibly measure attitudes toward technology and science. Among the miscellaneous preference items were "factual science articles" and "factual reports on the space program and spaceflight." These two are strongly associated (r = 0.57), as one might expect. Each is strongly correlated with Hard-Science and with fiction based on the physical sciences. But there is no significant correlation, positive or negative, with New-Wave or Sword-and-Sorcery.
|Factual Science Articles||Reports on the Space Program||Anti-Technology Scale|
|Fiction based on the physical sciences||0.53*||0.49*||-0.35*|
|Fiction based on the social sciences||0.07||0.11||0.02|
|Stories about magic||-0.05||-0.12||0.06|
The third column of coefficients in Figure 4 represents an antitechnology scale based on the six items taken from the study by Irene Taviss. I must emphasize that fans of different types of literature disagreed very little in their attitudes toward technology. To get any meaningful variation, I was forced to combine the six items using a technique called factor analysis - using the first factor of the two that emerged as an anti-technology scale - producing a very sensitive scale based on all the items except the one on atomic power which the procedure found to be an independent issue.
The table shows that fans who like Hard-Science are more apt than others to enjoy reading about real science and about spaceflight. I think these are valid indicators of real enthusiasm and support. Unfortunately, Taviss' item on spaceflight got such a high level of response from all types of fans, that it was impossible to calculate reliable correlations on this item alone. Without variation, we do not have a variable! The anti-technology scale is so sensitive that it may deceive us into thinking we have found strong relationships when the associations are really very weak. Thus, the table shows significant enthusiasm for science and spaceflight among Hard-Science fans compared with others, and weak rejection of anti-technology sentiments. But, of course, this merely represents an accentuation of the tendency of all science fiction fans to have these attitudes.
Indeed, we do not have evidence that fans of New-Wave or Sword-and-Sorcery are anti-science, anti-spaceflight or anti-technology. On the contrary, the lack of correlations shows they are average SF fans on these issues -- and average SF fans are more positive on spaceflight, at least, than are other groups in the population. Of course, the vast majority of casual SF readers do not attend conventions, and our respondents represent the extreme in dedication to the science fiction subculture. We have seen the ideological factions and the impact of SF for fanatics of the literature. We must look beyond this extreme group if we want to determine the influence science fiction may have on the wider culture.
The best data presently available come from an ongoing series of questionnaire studies performed at the University of 'Washington in Seattle. Other research has shown that the student body is quite representative of the values and interests of middle-class young Americans, even though slight regional differences do exist. If the results of the largest of these student surveys shows patterns at all similar to those found in the survey of SF fans, then we will be much more confident in our preliminary conclusions.
Among the items included in the student survey were the six Taviss statements, others about space-related issues, and a large section of preference items of the same form as those in the science fiction questionnaire. Students expressed opinions about several movies, TV shows, and kinds of literature which included "science fiction novels." This SF item was significantly correlated with preferences for movies "Star Wars" (r = 0.31) and "Close Encounters" (r = 0.25) and for TV shows "Battlestar Galactica" (r = 0.33) and "Star Trek" (r = 0.49). Therefore, we can use these five items as good indicators for science fiction in three media.
One of a number of agree-disagree opinion statements was: "The United States is spending too much money on space, so appropriations for the space program should be reduced." Overall, 31.5 percent of the students wanted space appropriations reduced. Many of those who disagreed undoubtedly felt that current expenditures were at the right level, so we cannot conclude that a majority wanted the space program strengthened. Figure 5 graphs responses to this statement against the degree to which students liked science fiction novels and the two television programs. "Star Wars" was too well liked, and "Close Encounters" too seldom seen, to include without greatly complicating the picture. Each of the three lines on the graph is defined by seven data points. Each data point represents the students who gave a particular rating to the given preference item, and its height on the graph indicates what percentage felt space appropriations should be reduced.
To examine student attitudes toward a wider range of opinion statements we must use a table, as in Figure 6. The first six items are Taviss' technology statements. In the surveys reported above, respondents were allowed to check a box labeled "not sure" if they could not make up their minds, an option not given to the 1,439 college students. Therefore, we should not compare percentages outside the student survey. Figure 6 also includes the space appropriations item and two about life and civilizations on other planets.
The first two columns compare opinions of those who dislike SF (rate it "2" or lower) with opinions of those who love it (rate it "6"). The third column in Figure 6 expresses the relationships through a coefficient called tau, more appropriate for these opinion items than Pearson's r. Tau will naturally give us somewhat smaller numbers than we have seen in previous tables. But magnitude of the coefficients aside, Figure 6 shows us just what we saw among diehard SF fans. There are weak negative associations between liking science fiction novels and expressing anti-technology statements - very weak and even statistically insignificant as measured by tau. There are no correlations with general faith in technology or support for atomic power. But there are really strong differences on the four items about space. SF readers are much more likely than other students to support the space program and to believe that contacting extraterrestrial civilizations is both possible and desirable.
But do these correlations really prove that science fiction produces pro-space attitudes? It could be that people who already favor spaceflight are drawn to read science fiction. A correlation coefficient, by itself, cannot tell us the direction of causation. Perhaps people who already favor spaceflight therefore come to prefer science fiction. Or perhaps some other variable lurks in the background, causing both love of space and love of SF. One thing we can do is hunt for such third variables and see if they can account for a "spurious" relationship between SF and space. Most of our students are young and middle-class, and there is little variation in their level of education, so three of the standard variables cannot explain differences in these attitudes. That leaves sex. Of the 1,439 students, 682 said they were women, and 741 admitted to being men. Traditionally, both science fiction and spaceflight have been male passions. Perhaps the apparent relationship between SF and pro-space attitudes is just a result of the fact that men happen to like both. The two concluding columns of the table look at correlations between SF and the nine opinion items, separating respondents by sex. The results are almost identical among women, among men, and in the total group. So we are left without standard variables known to influence attitudes toward space, which might be responsible for a spurious association between SF and support for the space program.
|Percent who agree||Correlation (tau) with SF novels|
|556 who dislike SF||198 who like SF||1,439 total||682 women||741 men|
|1. Machines have thrown too many people out of work.||24.5%||16.2%||-0.10*||-0.07||-0.06|
|2. It would be nice if we would stop building so many factories and go back to nature||42.9%||36.4%||-0.06||0.00||-0.02|
|3. Technology has made life too complicated.||26.1%||21.8%||-0.06||-0.06||-0.01|
|4. Technology does more good than harm.||65.1%||72.2%||0.08||0.03||0.07|
|5. The potential dangers of nuclear energy are outweighed by its potential benefits.||47.1%||53.3%||0.03||0.01||0.04|
|6. In the long run, discoveries made in our space program will have a big payoff for the average person.||50.3%||81.2%||0.25*||0.19*||0.26*|
|7. We should attempt to communicate with intelligent beings on other planets, perhaps using radio.||45.8%||77.0%||0.22*||0.22*||0.17*|
|8. Intelligent life probably does not exist on any planet but our own.||25.7%||6.6%||-0.24*||-0.20*||-0.21*|
|9. The United State is spending too much money on space, so appropriations for the space program should be reduced.||43.5%||14.3%||-0.24*||-0.17*||-0.25*|
There is another justification for concluding that science fiction does have the expected propagandistic influence. While not conclusive, the fact that other technology items have little or no correlation with SF suggests that there is no hidden global variable making some people simultaneously favor space and SF. If all the technology items went together in the correlations, then we might suspect there were major factors of family background, personality, education, or whatever that produced all the attitudes. But the very complexity of the picture makes me doubt that any such confounding variable lurks just under the surface.
Furthermore, science fiction is one of the optional personal pleasures offered people in their daily lives. They can choose to try it, and some of them will like it for complex idiosyncratic reasons. Consumption of much SF will change a person, will teach them to think of space as a good place, an environment filled with interesting possibilities worth exploring. It is much harder to imagine that spaceflight itself directly enters the lives of many people today. In a future age when millions go on tours to the moon, I can well imagine that such a trip sets them to reading SF after they get home. But this is not the situation today. All things considered, the most reasonable interpretation of the data is that science fiction strongly promotes spaceflight, while having only the weakest capacity to prevent anti-technology sentiments, and no power to produce favorable attitudes toward technology in general. We must conclude that while science fiction does support the space program, it teaches a much more questioning stance toward other forms of technological progress.
A final question is whether only science fiction literature has these effects, or if "sci-fi'' movies and TV shows effectively communicate pro-space values as well. Figure 7, our last, shows correlations (again the relatively weak-appearing tau) between the nine opinion items and all five science fiction preference items which were included in the college student survey. We see that there are differences. "Star Trek," the phenomenally popular TV show, has associations nearly as strong as does science fiction literature. The others have lower, but mostly significant correlations. In general, then, mass visual media do dilute the science fiction message. But still it gets through.
The reader may feel that this paper has merely belabored the obvious science fiction promotes spaceflight. But this conclusion is not at all obvious, for three reasons. First, one might have guessed that science fiction strongly promoted favorable attitudes toward all kinds of technology, but we have found that it has only a weak capacity to discourage boldly anti-technology opinions and no capacity to promote atomic power or a general faith in the goodness of technology. Second, we have seen that contemporary science fiction contains at least three competing ideological factions, two of which might logically oppose spaceflight rather than promote it. Finally, one might argue that people simply do not take fiction seriously, so they are not influenced by it - that SF is an escapist genre having no implications for the real world.
|Correlation (tau) with preference for|
|SF novels||Star Wars||Close Encounters||Star Trek||Battlestar Galactica|
|1. Machines have thrown too many people out of work.||-0.10*||-0.04||-0.01||-0.09*||-0.03|
|2. It would be nice if we would stop building so many factories and go back to nature||-0.06||-0.03||-0.03||-0.04||-0.05|
|3. Technology has made life too complicated.||-0.06||-0.04||-0.02||-0.05||-0.06|
|4. Technology does more good than harm.||0.08||0.06||0.05||0.06||0.04|
|5. The potential dangers of nuclear energy are outweighed by its potential benefits.||0.03||0.04||0.00||0.02||0.02|
|6. In the long run, discoveries made in our space program will have a big payoff for the average person.||0.25*||0.12*||0.12*||0.19*||0.15*|
|7. We should attempt to communicate with intelligent beings on other planets, perhaps using radio.||0.22*||0.10*||0.14*||0.20*||0.12*|
|8. Intelligent life probably does not exist on any planet but our own.||-0.24*||-0.09*||-0.10*||-0.22*||-0.09*|
|9. The United State is spending too much money on space, so appropriations for the space program should be reduced.||-0.24*||-0.10*||-0.08*||-0.16*||-0.15*|
Thus, it is a confirmation of an arguable hypothesis, if not a stunning discovery, that we find a strong positive association between liking science fiction and favoring spaceflight. The Hard-Science variety of SF appears to have some extra power, above and beyond that of other varieties, to encourage space enthusiasm. Contrary to what one might predict, New-Wave and Sword-and-Sorcery do not breed disillusionment with the space program. The emergence of the New Wave in the 1960s, and the explosion of interest in Sword-and-Sorcery in the 1970s, might cause worry that science fiction was losing its capacity to boost the space program. The results of our convention and student surveys dispel this fear and confirm that contemporary science fiction continues to serve the cause of interplanetary exploration.
1. Gernsback's personal ideology is revealed in his editorials, for example in the first issues of Amazing Stories (1926), Science Wonder Stories (1929) and Science-Fiction Plus (1953), and in the June and October 1926 issues of Amazing Stories.
2. William Sims Bainbridge, The Spaceflight Revolution (New York: Wiley-Interscience, 1976).
3. William Sims Bainbridge, Dimensions of Science Fiction (Cambridge, Massachusetts: Harvard University Press, 1986).
4. William Sims Bainbridge and Rodney Stark, "The 'Consciousness Reformation' Reconsidered," forthcoming in Journal for the Scientific Study of Religion; "Friendship, Religion and the Occult: A Network Study," forthcoming in Review of Religious Research.
5. Irene Taviss, "A Survey of Popular Attitudes Toward Technology," Technology and Culture 13:4 (October 1972): 606-621.
6. William Sims Bainbridge and Richard Wyckoff, "American Enthusiasm for Spaceflight," Analog 99:7 (July 1979): 59-72.