[See also short responses to this article by Finkelhor and Mrazek]
Summary: Three critiques of Theo Sandfort's research on man-boy sexual relationships in the Netherlands are examined and evaluated. Three types of criticism - methodological, speculative and moral - are identified. Specific criticisms of the study are evaluated on the basis of their validity and, where appropriate, their underlying assumptions. It is argued that moral condemnation of such relationships, combined with a prevailing ideology of boy "victims" and adult "perpetrators," results in efforts by Sandfort's critics to attack and discredit his research rather than evaluate it objectively.
In 1981 Theo Sandfort, lecturer in psychology at the State University of Utrecht in the Netherlands, published the resuls of his study of 25 boys aged 10 to 16 who were involved in ongoing pederastic relationships with adult men at the time of the study. He concluded that "For virtually all the boys ... the sexual contact itself was experienced positively and had no negative effect on how the youngster felt in general" (Sandfort, 1982, p.85). Sandfort also determined that the boys did not perceive a misuse of power by the men. An English translation of his book appared in 1982. In 1983, an article by Sandfort entitled "Pedophile relationships in the Netherlands: Alternative Lifestyles for Children?" appeared in the Spring 1983 issue of Alternative Lifestyles. This article was not intended as a full summary of Sandfort's study but instead, in the author's own words, concentrated on "the meaning of the older partner and the pedophile relationship itself for the younger partner" (Sandfort, 1983, p.165). In 1984, an article providing a fuller summary of Sandfort's research appeared in the "Journal of Sex Research" under the title "Sex in pedophiliac relationships: An empirical investigation among a nonrepresentative group of boys."
Since the publication of Sandfort's findings, his research has begun to receive some reaction from American writers and researchers. Among those who have offered criticism of his research are David Finkelhor, author of the 1979 study Sexual Abuse of Children and other articles; David Mrazek, co-editor of Sexually Abused Children and Their Families and author of several articles on the topic of child sexual abuse; and the world-famous sex researchers William Masters and Virginia Johnson, in a textbook written with Robert Kolodny (1985). Finkelhor's criticism appeared in a brief review of Sandfort's work in Forum magazine (1984); Mrazek's, in a review of Sandfort's book in Contemporary Psychology (1985). This paper will examine the types of criticism made by these authors and their validity. In addition, it will address the question of the ideological viewpoint from which they are made. The criticism of Sandfort's work can be divided into three major categories: methodological, specualtive, and moral. The moral criticisms, which are arguably the most revealing, will be discussed last.
One of the first methodological criticisms of Sandfort's study is that his findings are invalid because his sample was unrepresentative. This criticism is made by all three writers: Finkelhor, Mrazek, and Masters, Johnson, and Kolodny. This would be a strong point against the study if Sandfort had claimed that his sample was representative or if he had been trying to demonstrate that most boys experience such contacts positively. However, Sandfort readily admits in all his writings that the sample can in no way be called representative of all boys who have sexual contact with men. He carefully emphasizes the limitations of his research, and states that his conclusions cannot be generalized to all man-boy contacts, to contacts between men and girls, or to incest.
More importantly. Sandfort's critics fail to address the question of whether or not a representative sample was necessary for the actual research question. Sandfort writes: "The question was whether a sexual contact with an adult could be a positive experience for a child. To the extend to which this research material can give a definite answer, the question must be answered in the affirmative" (Sandfort, 1982, p.84). Whether or not the sample was representative is beside the point. Since the study only sought to address the question of whether such relationships could ever be experienced positively, an unrepresentative sample was adequate.
It is very disturbing to note the lack of any criticism of the unrepresentative samples used by other authors who claim that such relationships are always negative (see Adams-Tucker, 1982; Burgess, 1984; and Peters, 1976 for examples). Where were Sandfort's critics when studies using rape victims, psychiatric patients, criminal cases, and emergency room patients were used to draw conclusions about the nature of all adult-child sexual relationships? Evidently if a study claims negative findings then an unrepresentative sample is perfectly acceptable, but if it claims positive findings then an unrepresentative sample renders it invalid.
Mrazek elaborates further on the theme of biased sampling. After noting that the boys were contacted through self-identified pedophiles who belonged to the Netherlands Association for Sexual Reform, he argues that "one is faced with the reality that the study sample are all men actively involved in the pedophile movement who have strong self-interest in the results of the research." and later complains that "the motivation and selection of the sample" is a problem not sufficiently addressed (Mrazek, 1985, p.37). Mrazek seems a bit confused here - the study sample includes the boys, not hte men. The motivation of the men is irrelevant. It should also be noted that all interviews with the boys were conducted in private and the boys were assured that their partners would not be allowed to see their answers to the interview questions, clearly reducing the likelihood that the boy's responses were influenced by the older partner.
A second methodological criticism is that Sandfort failed to define "negative" effects on the boys and furthermore, according to Masters et al., that he "gave no evidence of having asked appropriate quesitions to discover if these effects were present or not" (Masters, Johnson, and Kolodny, 1985, p.450). Unlike the complaints of unrepresentativeness, this criticism is not irrelevant; it is simply false. As a part of the structure of his research, Sandfort presented each boy with a list of positive and negative emotions and behaviours. After it was established that each boy understood these emotions and behaviours, he was asked to indicate how often he experienced them in both his overall relationship with the adult and in his sexual contacts with the adult. The boy was given clear and precise means to show whether or not he had perceived his experiences negatively. The various correlations between how the boy felt in the relationship,, how he experienced other relationships and activities in the boy's life and how it affected his overall sense of well-being. In addition, Sandfort noted athat when the boys were asked to formulate negative aspects of the sexual contacts "it was often difficult for the boys to discover any negative sides, which made it necesary to question them rather insistently" (Sandfort, 1982, p.64). Far from not having asked "appropriate questions," Sandfort actually pressured the boys to come up with negative aspects.
In a similar criticism, Mrazek attacks Sandfort's study for failing to address the issue of whether or not these relationships might result in the boys' developing a sexual deviation "from more normal heterosexuality" (Mrazek, 1985, p.38). However, there is no evidence that such relationships result in the younger partner becoming a homosexual. Longitudinal studies of boys who were sexuallly involved with men during adolescense (Tindall, 1978) and sudies of adults who had such experiences as boys (Bernard, 1981) have failed to find such evidence. In addition, studies of societies where sexual relationships between men and boys, or between boys and boys, are institutionalized have not found any evidence of widespread adult homosexuality (money and Ehrhardt, 1972; Herdt, 1985). Extensive or even exclusive homosexual relations over a period of years in childhood and adolescense do not seem to prevent boys from becoming heterosexual adults.
Mrazek also complains that Sandfort fails to address the risk of promicous early homosexual contact, including a high incidence of genital herpes and increased risk of AIDS. It is true that Sandfort does not address this issue, but the criticism has only limited validity. Men who are pederasts typically do not have sexual contacts with other adult men. The high rate of herpes and the increased risk of AIDS do not apply to pederasts as a group as much as they apply to homosexuals (although AIDS is now becoming more widespread in the general population). There is a very real distinction between the two groups. In addition, there is no indication that either the men or the boys were engaging in "promiscous" homosexual contact; it seems to be a gratituous accusation thrown in by Mrazek. Finally, most of the sexual contacts involved the partners masturbating each other; this was followed in frequency by oral sex (most often performed on the boy by the man). Anal contacts occured in only a small minority of relationships. Since AIDS aeems to be transmitted mainly through exchange of body fluids in penetrative sexual contacts, most of the man-boy sexual contacts were of the "low-risk" type.
Master et al. offer a third methodological criticism: "Unlike standard practice in research of this sort, no psychological tests were administered to evaluate the boys' emotional stability or self-esteem..." (Masters, Johnson, and Kolodny, 1985, pp450-451). In the first place, research of this sort has been very scarce; and for that which exixts, it has not been standard practice to administer psychological tests. For example, the often-cited Finkelhor (1979) study of college students, which is regualrly invoked as "proof" that children experience such contacts negatively, reached this conclusion by questionaire only. The more recent study by Burgess et al. (1984) also reached its conclusion that all such relations are harmfull (actually, this "research" simply took harm as given) without the benefit of a single psychological test. In fact, those studies which have actually used psychological tests, such as those of Burton (1968) in England and Bauermann (1981) in West Germany, often come to the conclusion that such tests show no psychological damage form the sexual relationship or encounters. In the second place, Sandfort's study is a psychological evaluation. Sandfort employed an approach called the Self-Confrontation Method based on valuation theory. This method has been used with great success in other areas of personality research in the Netherlands and Sandfort simply applied it to the pederastic relationships. With this method, it was possible to determine what aspects of the boy's experience world were important to him; how he expewrienced these aspects (positively or negatively); how frequently he experienced particular feelings and behaviors, both positive and negative, in connection with each area; and how each area influenced his overall well-being and corresponded with how he would most like to feel. It is the subject (the boy) who interprets his experience, not the researcher. As to self-esteem, the comparison between the frequency of the feelings hte boys reported in the relationship with the man, and the frequency of the various feelings that the boys said they would like to experience in general, showed that the two correponded closely. The sexual relationships with the men had a positive overall effect on the boy's sense of well-being and helped bring them closer to feeling the way said they would most like to feel.
In the fourth criticism, Masters, Johnson, and Kolodny state that "Even more unbelievably, each boy was interviewed in the home of 'his' pedophile wit hthe pedophile present, without any apparent regard for the fact that the adult's presence would have almost assuredly prevented the boy from voicing complaints about the way he was treated because of fear of punishment" (Masters et al., 1985, p.451). It is this criticism that is unbelievable. This statement is either a knowing lie, if Masters et al. read Sandfort's actual study, or an irresponsible assumption, if they did not. Sandfort stated in his book in at least three different places that the interviews with the boys were conducted in private with only the boy and the researcher present, in order to prevent the older partner from influencing the boy's answers through his presence. Sandfort also makes this point in his article for the Journal of Sex Research. Either Masters et al. did not bother to read Sandfort's acutal study (in which case they have no business critizing it) or they did, and chose to print this claim anyway.
A check of the bibliography for Masters et al.'s textbook reveals that the only work of Sandfort listed is his 1983 article for Alternative Lifestyles. Since neither Sandfort's book nor his article for the Journal of Sex Research are listed in their bibliography, it seems that Master's et al. did not read either of these. As mentioned above, Sandfort did not intend this article as a comprehensive summary of his findings but rather as an effort to understand the meaning of the older partner and the relationship for the younger partner. In this article, Sadnfort does not clearly state that the interviews with the boys were conducted privately. He also does not imply that the older partner was present during the interview. It seems that instead of referring to Sandfort's book or to his article in the Journal of Sex Research in order to be sure themselves, Masters et al. chose to assume that the older partner was present and to print this as a fact in an apparent effort to discredit Sandfort's research. The fact that they made such an assumption and then failed to make any effort to check its accuracy suggests that they are more concerned with discrediting Sandfort's study than with making valid and responsible criticisms of actual flaws the study might possess.
Masters et al.'s last criticism of Sandfort's methods was that "no follow-up of the boys and their relationships was attempted to discover what long-term impacts might be" (Masters et al., 1985, p.451). Sandfort noted in his study that "considering the short period of time elapsed during this research, it was not possible to examine the long-range consequences of the sexual contacts" (Sandfort, 1982, p.84). Since the original research was carried out in 1980 and all of Sandfort's writings appeared within the next four years, it is obvious that not enough time had elapsed since the original study for a meaningful long-term follow-up to take place. Significantly, Sandfort also writes that all the boys agreed to take part in a long-term follow-up. In light of these facts this criticism must be seen as an irrelevant accusation by Masters, Johnson and Kolodny with the apparent intent of smearing Sandfort's scientific integrity and the credibility of his research. It should also be noted that one of the major assets of Sadnfort's study is that it deals with boys who at the time were currently involved in ongoing sexual relationships with an adult. The sample thereby avoided the problems of adult recall studies and of studies of boys whose relationships had been ended by police or other social intervention.
One of Mrazek's major criticisms is the terminology of the study. He states that the usual labels of "perpetrator" and "victim" are "militantly avoided" and later states that "the reader who believes that young children should be preotected form sexual advances of adults will find the language with which the 'researcher' describes these relationships offensive" (Mrazek, 1985, p.37). Examples of the offending terminology include using the term "partners" for the man and boy instead of "victim" and "perpetrator," and the use of the term "making love" for the sexual contacts.
This criticism seems to stem from a conviction that all adult-juvenile sexual contacts are by definition abusive and exploitative. However, avoiding the terms victim and perpetrator is not militant. "Partners" is a relative neutral term that does not carry the value-laden negative connotations of "victim" and "perpetrator" nor the equally value-laden positive connotations of the term "lover." Other writers besides Sandfort have pointed that the offender-victim dichotomy is not always appropriate (Schultz, 1973). The fact that some boys see themselves as consenting participants has led to the coining of the phrase "participating victim" in an apparent effort to maintain the offender-victim distinction. As to term "making love," Mrazek fails to point out that it is the boys themselves who refer to the sexual contacts in this fashion - not the researcher. One of the questions Sadnfort asked of the boys was the term they used for having sex with their adult partner, and a number of them referred to it as "making love."
A final methodological criticism also comes from Mrazek's review. He argues that while Sandfort initially acknowledges his sampling bias and warns that his findings cannot be generalized, a "logical contortion" is necessary to reach his conclusion that "in these cases the basis which justified the criminalizing of such contacts simply doen's exist, and the result of this research suggests no other justification for prohibiting these sexual contacts." Where is the logical contortion? Sadnfort simply points out that for consensual contacts of the sort that he studied, the basis for criminalizing such contacts - presumend harm to the younger partner - does not apply. Suggesting that relationships which a boy regards as consensual and positive should be decriminalized in no way conflicts with continued protection of children form sexual contacts brought about by force, coercion, deceit, or other manipulation by and adult.
Several specualtive criticisms are also offered by Finkelhor and by Masters, Johnson, and Kolodny. These criticisms do not focus on the methodology of Sandfort's study, but rather conjecture about possible reasons ofor rejecting his findings. Finkelhor questions whether or not boys were honest with the researcher. He states that "children" were selected by a group of sensitive and intellectual pedophiles and were therefore unusual. However, he offers no explanation as to why this might make the boys be dishonest with the researcher. What seems to be implied here is that the boys presented a falsely positive picture of the relationship. But why would the boys want to portray a negative relationship as a positive one, when they presumably the ones being "victimized?" Furthermore, how could the boys possibly lie so consistently and thoroughly as to leave no indication of a negative attitude towards their relationship with the older partner or of an overall negative impact of the relationship; the behaviors they encountered in the relationship; their relations with parents, friends, and others; and their attitudes toward home, school, hobbies, etc? The suggestion that the boys were dishonest is pure speculation and seems unlikely at best.
Finkelhor goes on to state that his own research shows that "most kids react negatively" to sexual relations with an adult. In point of fact, the majority of males in Finkelhor's study rated their sexual relation with older males as positive or neutral; less than 40% said their encounters, as boys, had been a negative experience. Furthermore, Finkelhor fails to mention the inherent bias in his own study, which used a loaded questionaire seemingly designed to preclude the possibility of reporting consensual, nonincestuous relationships with adults. Finkelhor's questionaire, while asking subjects about all experiences before age 12 with older partners, only asked subjects to report incestuous or nonconsensual experiences that occured after age 12. If any of Sandfort's subjects had taken this questionaire, they would not have reported their positively experienced relationships since these did not fall into either category.
Finkelhors final criticism in this category is that there is such a fundamental asymmetry of power between adult and "child" that kids have no real power of consent. However, many boys clearly view themselves as willing participants and do not feel that they have been manipulated or coerced. Even more significantly, adults reflecting back on their relationship as children repeat this point of view (as the studies by Bernard, 1981; Ingram, 1981; and Tindall, 1978, demonstrate). The boys in Sandforts study were aware of the possible negative reactions of parents, friends, and authorities to the relationship, yet chose to continue it anyway. In light of these facts it seems hard to maintain the belief that boys are somehow incapable of saying no.
Masters, Johnson and Kolodny speculate that perhaps the "boys were so intimidated by the pedophile that they were afraid to say anything against him" (Masters et al, 1985, p.451). It should be noted that a number of studies have found that there are boys who regard their relationships as having been positive even after they became adults; and in addition, many maintain close ties with their older partners well into adulthood (see Bernard, 1981; Ingram, 1981; Rossman, 1976; and Tindall, 1978, for examples). This is hardly the attitude one would expect from someone who was intimidated by, and scared of, their older partner. Finally, the boys in Sandfort's study did say some things against their older partners. They felt they did encounter some negative behaviours such as coercion, although only rarely, and also reported that sometimes they behaved negatively towards their older partners as well. In other words, these man-boy relationships were described by the boys just as any sort of positive relationship between two people might be described. Although positive behaviors predominated, petty fighting, manipulation, and coercion on the part of both partners occured as well. If the boys had claimed that they always and only experienced positive behaviour, one would have every right to suspect dishonesty, but they did not. The fact that the boys reported both good and bad feelings and behaviours in the relationships indicates that they were being honest when they reported that the good far outweighed the bad in their eyes. Masters et al.'s specualtion that the boys were intimidated is not based on any interviews with the boys in question. In light of the facts from Sandfort's own study and the other lkiterature cited above, this specualtion is unwarranted.
Masters et al. conclude their criticism by explaining why adult-child sexual encounters are always negative. They state that they agree with Dr. Suzanne Sgroi, co-director of the St.Joseph College Institute for the Treatment and Control of Child Sexual Abuse, who says "... The sexually abused child may not feel abused initially, but as the child learns what society thinks of what he has done, the child feels betrayed. He feels he cannot trust adults or family members ... [and] has a sense of danger, of being violated, a sense that he is not as good as he was before" (Masters et al., 1985, p.451). If Dr. Sgroi's analysis is correct, it only shows that society's reaction makes the child feel abused, not the sexual contact itself. In the past (and even today, to a lesser extent), society's attitudes toward sexual behaviors such as masturbation and homosexuality caused untold guilt, shame and anxiety. It is now widely recognized that the "harm" attributed to masturbation and homosexuality came more often than not from the societal reaction to these behaviors. In fact, it is now well understood that our society has tended to react with horror to many types of harmless sexual behaviors. To base a blanket condemnation of all adult-juvenile sexual encounters on society's reaction is erroneous.
It should also be noted that Dr. Sgroi's analysis (at least as far as boys are concerned) is not supported by the literature. The studies of Bernard, Ingram, and Tindall show that many boys, far from feeling betrayed and unable to trust people, continue close friendships with their older partners well into adulthood and show no evidence for social or psychlogical maladjustment. In addition, as statet earlier, Sandfort found in his study that the boys were well aware of negative societal attitudes and the possible negative reactions of parents, friends, and authorities. However, they still viewed the relationships as positive and felt no sense of betrayal or lowered self-esteem. All of this directly contradicts Dr. Sgroi's contentions.
The final category of criticism consists of moral criticisms. All three critics of Sandfort's study include moral criticism of one sort or another. As stated above, it is the moral attacks that may provide some of the best inside into the ideological viewpoints held by Sandfort's critics. Finkelhor claims that Sandfort's findings do not alter the "moral perspective" of the issue and by way of analogy says that "there were probably slaves who loved being slaves and were not hurt by it, but society determined that the institution was a violation of fundamental ethical standards and abolished it" (Finkelhor, 1984, p.8). It is hard to see how this related to the relationships studied by Sandfort. While a slave is completely within the power of the slaveowner and is denied basic and fundamental freedoms, there is no evidence that the boys in question felt compelled or forced to continue the sexual relationship. In fact, most of the boys took the initiative in sexual contact with the adult partner at least from time to time. Some explicitly stated that the relationship were a matter of thier personal choise and said they felt they had the right to make decisions for themselves about whether they participated in a sexual relationship or not. Finkelhor's analogy seems based on the assumption that such relationships are automatically wrong because of power differences, regardless of the boy's attitude toward the relationship, its impact on the boy, and whether or not he felt coerced or forced.
A more serious moral criticism comes from Mrazek's review. He argues that since the sexual contacts in question were illegal in the Nehterlands, Sandfort was rationalizing criminal behaviour and was maintaining the secrecy of a sexual offense from the boys' families. While this may posean ethical dilemma, it should be pointed out that many other studies of sexual behavior are open to the same criticism. For example, Kinsey interviewed many adolescents and prepubescent boys about their sexual behavior for his work Sexual Behavior in the Human Male (1948), recognized as a landmark in sex research. However, sexual behavior, while legally acceptable in adults, is ground for legally judging an adolescent or child a delinquent. Kinsey himself pointed out that from a specific calculation from his data, some 86% of adolescent boys in America were sexual delinquents. Should Kinsey have repoteed all these boys to the authorities as sexual delinquents? Later researchers into adolescent sexual behavior, such as Sorensen (1973), can be attacked on the same grounds. Researchers investigating homosexuality in the 60s and 70s are also open to the same criticism - nearly all states in the U.S had laws against homosexual behaviour in the 1960s and many still have today. Should these researchers have turned their subjects in to the police for prosecution because they were violating tha law by participating in homosexual encounters with other men? What about psychiatrists and psychologists with homosexual patients? Should they have reported their patients to the police for prosecution?
This ethical dilemma is by no meas an easy one to resolve. However, Sandfort was faced with a group of boys who statet that they enjoyed their relationship with the older partner and the sexual contacts involved. In addition, in 7 of the 25 cases the boy's parents were aware of the relationship and accepted it. Would it have helped these boys in the short or the long term if their relationships had been reported to the police and broken up? Mrazek also fails to note that a very different climate exists in the Netherlands than in the U.S. regarding adult-child sex. Police are reluctant to prosecute when there is no evidence of coercion and when the youth involved is at least 12 years old. In addition, an ongoing public debate on the age of consent has even included appearances by children and their adult partners on national radio to discuss their relationships. In such a climate, the issue of reporting does not carry the same weight as it does in the U.S., where professionals are required by law in many states to report such relationships. It should also be noted that in the Netherlands, clinicians are not under any legal obligation to report crimes committed by their clients ( with the exception of assaults on someone's life), unlike the situation in many states in the U.S.
The inclusion of a moral perspective to condemn man-boy sex is perhaps most surprising in the case of the textbook by Masters, Johnson and Kolodny. Earlier in their text, Masters et al. pointed to the misconseptions engendered and maintained by moral arguments. For instance, we are informed that th Catholic Church continues to insist that "masturbation is an intrinsically and serioously disordered act" and that this "unnatural act" came to be described as "self-abuse", "Defilement of the flesh," and "self-pollution." In the case of homosexuality, Masters at al. tell us that this form of sexual activity has been described in judicial decisions from the 1970s as "loathsome and disgusting," " grossly repugnant," "degenerate," "foul," and "immoral, lewd, and obscene." They quote one psychiatrist who, in 1970, described homosexuality as "a dread dysfunction, malignant in character, which has risen to epidemic proportions" (Masters et al., 1985, p.407).
Thus, it is surprising that Masters et al. ask in reference to man-boy relationships, "is an inherently abusive exploitive relationship 'positive' under any circumstances?" (Masters et al., 1985, p.451). They are aware of how such loaded terminology and sweeping condemnations of whole categories of sexual behaviour prevented an objective understanding of masturbation and homosexuality, and they are aware that a clearer underatanding was attained only through a deliberate disregard for such moral condemnationin favour of concentration on empirical research. But ignoring this lesson in the case of their own evaluation of man-boy sex, Masters et al. have strengthened the impression that science is only welcome so long as it fits in with the prevailing ideologies. Tehy are evidently perfectly comfortable in condemning all man-boy sexual relationships as "inherently abusive and exploitive," even though they reject this sort of claim as it was employed in the past against other types of sexual behavior.
The moral criticism of Sandfort's work thus appear to all be guided by a particular set of beliefs about adult-juvenile sex. These beliefs center primarily on the idea tha such a relationship are by their nature abusive and exploitive. They include a belief that the younger is automatically incapable of consent, because of the adult's position of greater power; a belief that the adult inevitably exploits the younger partner for his own needs, without concern for the youngster's own feelings and desires; and a belief that such relationship always or nearly always have negative consequences for the younger partner. More subtly, but just as significantly, they also seem to be based at least in part on a belief that deviations from the adult heterosexual norm are automatically suspect (witness Mrazek's comment on "more normal heterosexuality").
The methodological criticisms of Sandfort's work are variopusly distorted, irrelevant, or just plain false; the speculative criticisms are either irrelevant or else biased by their complete failure to consider contrary findings in the literature; and the moral criticism ahow a dogmatic adherence to the belief that all adult-juvenile sexual relations are by definition abusive, exploitive, and harmful. It seems that the taboo against juvenile sexuality and particularly against adult-juvenile sex is still so strong that research which fails to support the prevailing ideology must be attacked and discredited, regardless of its actual validity. It remains to be seen whether scientific objectivity can prevail against the need to defend the current dogma on man-boy sexual contacts in particular and adult-juvenile sex in general.