Opinion about neonatal male circumcision is deeply divided. Some take it to be a prophylactic measure with unequivocal and significant health benefits, while others consider it a form of child abuse. We argue against both these polar views. In doing so, we discuss whether circumcision constitutes bodily mutilation, whether the absence of the child's informed consent makes it wrong, the nature and strength of the evidence regarding medical harms and benefits, and what moral weight cultural considerations have. We conclude that nontherapeutic circumcision of infant boys is a suitable mat-ter for parental discretion.
Routine neonatal male circumcision has been the subject of considerable debate among medical professionals. This subject, however, has received negligible attention in the bioethics literature. This suggests that most scholars working in bioethics do not consider neonatal male circumcision, unlike the practices of female genital excision that are common in parts of Africa and elsewhere, to be a morally troubling surgical procedure.1 This attitude toward neonatal male circumcision seems to be shared by many people, even in societies where male circumcision is performed infrequently. That is to say, relatively few people think that the practice is morally unacceptable, even if they them-selves would not have their sons circumcised.
But some people do consider routine circumcision of children (who are too young to consent) to be morally wrong or at least morally suspect. They take circumcision to be a severely injurious practice. For many of them any routine alteration of infant genitalia is a form a child abuse. By contrast, the advocates of routine neonatal circumcision believe that there are significant health advantages to circumcision and that these unequivocally override the costs and risks, which they believe are negligible. Although these are polar views, they are not infrequently expressed. We believe that both views are mistaken, and we shall argue to this effect.
Circumcision, the removal of the foreskin, can be performed at any age and for a variety of reasons. However, we shall focus primarily on circumcision of minors for nontherapeutic purposes. An adult's decision to undergo circumcision is uncontroversial. Even the most ardent opponents of circumcision have no opposition to an autonomous adult's choosing to be circumcised.2 Similarly uncontroversial is therapeutic (rather than allegedly prophylactic) circumcision. As most nontherapeutic circumcision is performed in the neonatal period, it is circumcision in this period that will be our primary focus, although much of our discussion will also be relevant to the circumcision of older children.
Most of those writing on the topic say that they would like parents to be able to make an informed decision about whether to have their sons circumcised. However, it is extremely difficult for those who have not immersed themselves in the literature on this topic to be able to make an informed judgment. This is, in part, because of the vastly differing interpretations of the evidence offered by different reviewers. The issues are further clouded by the use of emotive language by a number of authors, especially among those opposed to circumcision.3 We cannot hope to review everything that has been written on the topic. Nor can we comment on the methodology and quality of every study. We shall certainly not be offering new empirical evidence. What we do plan to do, however, is to clarify some conceptual issues, something that is often neglected in the debate about circumcision. We shall also offer what we hope is a balanced outline of the evidence before suggesting how it might reasonably be interpreted.
Those who believe that circumcision of minors is morally prohibited often suggest that removing the foreskin constitutes mutilation of a child. For instance, Denniston, Hodges, and Milos (1999) note that Stedman's Medical Dictionary defines mutilation as "[d]isfigurement or injury by removal or destruction of any conspicuous or essential part of the body." Male circumcision, they say, is the injurious and appearance-altering removal of a conspicuous body part and thus unquestionably constitutes mutilation. But this sort of argument begs the question. It assumes that circumcision disfigures and injures. Yet this is exactly what is in dispute in debates about whether circumcision constitutes mutilation.
This can be seen if we consider other surgical procedures such as breast reduction, liposuction, and rhinoplasty. These are all procedures that alter the appearance of parts of a person. Those who re-quest such procedures do not take them to be disfiguring. Similarly, those who circumcise their sons do not take removal of the foreskin to be disfiguring. Even if people can be mistaken about what constitutes disfigurement, it is still true that one cannot assume that a surgical procedure is disfiguring simply because it alters the body. It may be enhancing or it may be (aesthetically) neutral - neither disfiguring nor enhancing.
Of course, even nondisfiguring surgical procedures can be injurious. Again, however, not every surgical procedure, even one that removes healthy tissue, can be assumed to be injurious. That a surgical procedure is harmful is something that must be demonstrated rather than merely asserted.
It is also possible for a disfiguring surgical procedure, all things considered, to be beneficial rather than injurious. For instance, amputating a gangrenous leg is considered by most people to be disfiguring. Such people could term it a mutilation. However, if amputation were the only way to save a person's life, it would usually be beneficial. Where a mutilation is, all things considered, a benefit, it can be morally justifiable. Thus, even if circumcision is a mutilation, it does not inevitably follow that it is morally unacceptable. Further argument would be required to establish that conclusion. Although nobody would suggest that circumcision can save a life as directly as can amputation of a gangrenous leg, it is also the case that circumcision, if a disfigurement at all, is a much less radical disfigurement than a limb amputation. The benefit it would have to produce in order to be justified would thus need to be much smaller.
In short, then, whether circumcision is a mutilation and, if it is, whether it is an unacceptable mutilation can be established only by argument and not by mere assumption. Potential harms and benefits must be examined and weighed against one another. This stands in contrast to the view of some opponents of circumcision who, though they believe circumcision to be harmful, say that they would still be opposed to circumcision even if it were not (Denniston 1999). They take the mere removal of healthy tissue from a child to be sufficient grounds for condemning all nontherapeutic circumcision of boys (Denniston 1999). But the ethics of a surgical procedure cannot be assessed independently of whatever harms and benefits it does or does not have. To think that a moral judgment can be made without considering these is to adopt what sounds like a dogma rather than a reasoned conclusion.
One reason why some opponents of circumcision think that circumcision can be condemned without considering what harms or benefits it might have is that they think appropriate consent cannot be obtained. Children, they correctly note, lack the capacity to consent to circumcision. It is usually parents who provide consent for the circumcision of their children. However, those opposed to any nontherapeutic circumcision of minors claim that parents are entitled to consent to surgical procedures for their children only when medical necessity is immediate and clear (Denniston 1999; Van Howe 1997). Accordingly, they regard circumcision of children to be a form of assault - which is what surgery amounts to when appropriate consent or exceptional circumstances (such as necessity) are absent.
But is it really true that parents are morally entitled to authorize medical interventions only for clear and immediate medical necessity? In parts of the world where diseases against which children are often vaccinated are now uncommon, the necessity of such vaccination for any individual child is neither clear4 nor immediate. Moreover, there are very small but real risks (including death) from vaccination. A child's informed consent for such vaccination cannot be obtained. The choice (where government allows a choice rather than simply requiring universal inoculation) can thus either be deferred to proxy decision makers such as the parents or left to the adult that the child will become. But delaying vaccination can undermine much of its benefit. It thus seems entirely reasonable that parents or other guardians of a child's best interests be morally entitled to decide for the child. The role of a parent is not simply to save children from immediate catastrophe, but is to protect and foster a child's long-term best interests. That is why most people think that parents may consent on behalf of their children not only to vaccination but also to such procedures as orthodontics and various non-medical interventions, including schooling.
There are limits, of course, on the sorts of things to which parents can consent on behalf of their children. Typically the things to which parents may not consent are those that are unequivocally harmful to their children. Perhaps those op-posed to circumcision believe that it is just such a procedure. That, however, is a much stronger claim than the claim that circumcision is not a clear and immediate medical necessity, and it is accordingly much harder to defend. Nor can it be argued that nothing is lost by delaying a choice about circumcision of one's child until he can make it himself. This is because there are costs to delaying circumcision until adulthood. At the very least, circumcision may be psychologically unpleasant in adults in a way that it is not in infants. Moreover, the risks are greater in adults. Finally, although, as we shall show, the evidence for beneficial effects of circumcision is controversial, insofar as there are these benefits, they are significantly reduced if the circumcision is performed later in life.
Alleged Costs and Benefits
Having established that a moral assessment of neonatal circumcision cannot be made without considering whatever costs and benefits it may have, we turn now to the empirical evidence about these matters. We shall examine each issue individually to determine whether there is a cost or a benefit and, if so, its magnitude. After we have considered each of these issues, we shall attempt to draw an overall conclusion about the net medical value of neonatal circumcision.
A compelling objection to neonatal circumcision is that it has usually been practised without any anesthesia.5 It is now well known that neonates are capable of feeling pain,6 and it is widely accepted that neonatal circumcision is a painful procedure. This recognition has led to a search for suitable analgesia. Among the techniques that have been employed are topical analgesia with EMLA cream, dorsal penile nerve block, and ring block. In randomized controlled trials, the topical cream has been shown to be effective in reducing the pain response (Benini et al. 1993) but not as effective as the two nerve block techniques (Lander et al. 1997.)7 The high efficacy of dorsal penile nerve block has been repeatedly demonstrated in neonates (Kirya and Werthmann 1978; Williamson and Williamson 1983; Stang et al. 1988; Spencer et al. 1992; Arnett, Jones, and Horger 1990). Studies in adults have shown that either ring block or dorsal penile nerve block combined with anesthesia of the frenulum are more often effective than dorsal penile nerve block alone (Szmuk et al. 1994). The limited use of regional anesthesia is attributable to misperceptions that the procedure is difficult to perform, that it carries significant risks and that it causes more pain than the circumcision itself. In fact, the analgesia can be administered with ease. Although minor complications such as limited bruising have commonly been observed, these healed spontaneously (Poma 1980; William-son and Williamson 1983; Stang et al. 1988; Snellman and Stang 1995; Lander et al. 1997), and complications of any clinical significance are rare (Sara and Lowry 1985). Finally, the administration of the injections themselves has not been found to elicit a pain response (Kirya and Werthmann 1978; Williamson and Williamson 1983; Stang et al. 1988).
Even if adequate analgesia is provided for the procedure itself, concern might be raised about post-operative pain. We are not aware of any studies on such pain and its control in neonates. There seems to be no reason, however, why simple topical or systemic analgesics should not suffice. Therefore, concerns about post-operative pain cannot constitute strong grounds against performing the procedure.
As with any surgical procedure, circumcision carries a risk of complications, including most commonly bleeding and sepsis. Studies that have looked at large numbers of children who were circumcised have reported varying rates of complications from 0.06% (Speert 1953) to 55% (Patel 1966). This apparent discrepancy is attributable to the definition of complication that is employed. Those studies that report high complication rates have included even minor post-procedure oozing of blood from the wound. That interpretation, however, is unreasonably broad and is inconsistent with what would constitute a complication in any other surgical procedure. The consensus, even among those reporting high complication rates, is that the incidence of clinically significant complications is very low. It is commonly thought to be around 0.19% to 1.5% (Gee and Ansell 1976; Wiswell and Geschke 1989; Fredman 1969). It is also agreed that even where there are complications, the majority of these either resolve spontaneously or are easily resolved by simple medical intervention. There are, of course, instances of more severe complications, as described in some case reports. These can include denuded penile shaft, laceration or necrosis of the glans, urethral fistula, and death. However, these are very uncommon. The risk of death, for example, is less than 1 per 500,000 (Speert 1953).
It has often been claimed that circumcision is protective against penile cancer. A simplistic approach to this issue is to compare the incidence of the disease in societies where circumcision is widely practiced with its incidence in societies where only a minority of males are circumcised. Both advocates and opponents of neonatal circumcision have adopted this approach in (partial) support of their respective views. Opponents of circumcision, for instance, have noted that the incidence of penile cancer in the United States, where the vast majority of males are circumcised, is higher (0.9â€“1.0 per 100,000 males) than in Denmark (0.82 per 100,000 males) where circumcision is extremely uncommon (Denniston 1999).8 By contrast, proponents of routine circumcision cite the extremely low incidence of penile cancer in Israel (0.1 per 100,000 males), where circumcision is even more prevalent than in the United States (Schoen et al. 2000).
This is an indirect approach to the issue. It
- determines the incidence of a disease in two populations;
- notes the prevalence of circumcision in these populations; and then
- makes an inference about the relationship between the disease and circumcision status.
It is indirect because it does not actually determine whether patients with the disease are circumcised.
There are studies that have adopted a more direct approach. They have examined patients with penile cancer and established what proportion of them are circumcised. These include studies of penile cancer in New York (120 cases; Dean 1935), Illinois (139 cases; Lenowitz and Graham 1946), New York (100 cases; Hardner et al. 1972), and Michigan (156 cases; Dagher, Selzer, and Lapides 1973). None of these 515 patients were neonatally circumcised. One concern with these studies is that the investigators did not control for possible confounding variables, which might include smoking, sexual behavior, socioeconomic status, and/or sexually transmitted diseases. Nevertheless, the overwhelming nature of the results suggests that they cannot therefore be dismissed. More recent studies have attempted to control for potential confounding factors. For instance, a study of 110 penile cancer patients (and 355 controls) in Washington state and the province of British Columbia found that, although other factors might also increase the risk of penile cancer, not being circumcised neonatally carried a 3.2-times greater risk for the development of this disease (Maden et al. 1993).
Other recent studies have been even more refined, not only controlling for potentially con-founding variables but also distinguishing between various forms of (squamous cell) penile cancer. The results of studies that make this distinction suggest that circumcision is protective against the more but not less severe forms of this disease. The spectrum of diseases that can be included un-der this rubric ranges, in increasing order of severity, from penile intraepithelial neoplasia (PIN) to carcinoma in situ (CIS) to invasive penile carcinoma (IPC). A study published in 1993 from the Mayo Clinic in Minnesota of 34 patients with penile cancer found that although low-grade PIN occurred in the 12 neonatally circumcised men, all the cases of CIS and IPC occurred in men who were not circumcised in infancy (Malek et al. 1993). In an even more recent study of 213 penile cancer patients in California, in which the circumcision status was known in 207 of them, 84.3% of the CIS and 97.7% of the IPC patients were not neonatally circumcised (Schoen et al. 2000).
None of these studies can be regarded as ideal or definitive. Among some of their potential short-comings are their retrospective nature, small sample size, and use of self-report as a means of determining circumcision status.9 Nevertheless, the preponderance of evidence suggests that neonatal circumcision is protective against (or at least associated with a lower incidence of) the more severe forms of penile cancer.10 Given that it is the more severe forms of this disease that entail greater morbidity and mortality, preventing these is of greater benefit.
We are not claiming that penile cancer, or even its more severe forms, does not occur in men who were neonatally circumcised. Nor do we wish to enter into the (distracting) protracted debate about exactly how many cases (or case reports) of penile cancer there have been in U.S. men circumcised in infancy. Some proponents of circumcision have claimed that there have been only ten such case reports within the last 55 years (Schoen 1991; Schoen 1996; Schoen et al. 2000). Opponents of circumcision have criticized this claim at great length (Denniston 1999), citing a few additional case reports and noting that not every case will have been reported in the medical literature. While there might be technical validity to at least some of these criticisms, they fail to get to the heart of the issue. It is inadequate to cite a few case reports in response to the numerous sizable case series that have shown that the overwhelming majority of cases of invasive penile cancer occur in men who were not circumcised in infancy.11
Urinary Tract Infection
The relationship between urinary tract infection (UTI) and circumcision has been the subject of many studies, and the consensus in the medical literature is that circumcision is associated with a lower incidence of UTI (Ginsberg and McCracken 1982; Wiswell, Smith, and Bass 1985; Wiswell and Roscelli 1986; Wiswell et al. 1987; Herzog 1989; Wiswell and Hachey 1993; To et al. 1998). We are aware of no studies that have shown either the reverse or no association. Lower incidence of UTI is one of the benefits most commonly cited in support of circumcision (American Academy of Pediatrics Task Force on Circumcision 1989; Schoen 1990; Ganiats et al. 1991; Lawler, Bisonni, and Holtgrave 1991; Niku, Stock, and Kaplan 1995; Moses, Bailey, and Ronald, 1998; Schoen, Wiswell, and Moses, 2000).12
That said, it must be emphasised, however, that studies do disagree about the magnitude of the increased risk of UTI among uncircumcised boys. Some studies have reported a 3.7-fold (To et al. 1998) increased risk of UTI in uncircumcised compared with circumcised children. Others have suggested as much as a 12-fold increase (Wiswell and Hachey 1993).13
There have been criticisms of the methodology of these studies. These include the retrospective nature of most of the studies, the inclusion of primarily only hospitalized patients, and the failure to control adequately for confounding variables (such as socioeconomic factors). All these are legitimate concerns and do show ways in which the studies have not been ideal. However, it is not reasonable, for these reasons, to completely disregard these data. First, one cannot ignore that, despite the varying limitations of the numerous studies, all have pointed in the same direction. Second, not all available medical evidence conforms to the highest standards, and it is necessary to base practice on the available evidence, which, although defective, is not so thoroughly flawed as to be entirely use-less. Medical evidence does vary in its quality, and if one were to reject all evidence that was less than ideal, one would be left with no basis for decision making in many areas of medical practice, not to mention everyday life.
Having established an increased risk of UTI among uncircumcised boys, the significance of this risk needs to be assessed. Here, some advocates of circumcision have tended to overrate the value of circumcision. Any fair assessment of its significance must consider the following: While it is true that circumcision confers a 10-fold risk reduction of UTI, the absolute incidence of UTI is low, with 0.15% of circumcised and 1.5% of uncircumcised male infants developing such an infection. Put an-other way, UTI does not occur in 99.85% of circumcised infant males and in 98.5% of un-circumcised infant boys. Moreover, most UTIs occur in the first year of life and are easily diagnosed and treatable, with low morbidity and mortality. More serious complications of UTI, such as vesicouretic reflux, renal scarring, pyelonephritis, and renal failure are possible but occur with low frequency (Littlewood 1972).14 In summary, then, circumcision does seem to confer a small but real benefit in terms of UTI prevention.
Sexually Transmitted Diseases
Circumcision has also been claimed to be protective against some sexually transmitted diseases (STDs), including human immunodeficiency virus (HIV). Although the evidence for this claim is not unproblematic, the data on HIV is more consistent than that on other STDs. Furthermore, given the availability of a number of systematic reviews, as well as a recent meta-analysis on HIV and circumcision status, the data on HIV is easier to assess. For this reason, we shall discuss HIV and non-HIV STDs separately. We shall consider the latter first.
STDs Other than HIV
Genital ulcer disease (GUD) is one category of STD. Of the five studies that looked at syphilis, four reported reduced risk in those who are circumcised (Wilson 1947; Parker et al. 1983; Newell et al. 1993; Cook, Koutski, and Holmes 1994), and one reported no difference (Lauman, Masi, and Zuckerman 1997). Herpes was investigated in six studies. Three concluded that circumcised men have a reduced risk (Parker and Banatvala 1967; Taylor and Rodin 1975; Parker et al. 1983), and three found no difference (Cook, Koutski, and Holmes 1994; Donovan, Bassett, and Bodsworth 1994; Lauman, Masi, and Zuckerman 1997). One study of chancroid (Haemophilus ducreyi) found that circumcised men were less susceptible (Simonsen et al. 1988). Two studies examined genital ulcer disease without specifying the infecting organism, and both found reduced risk in circumcised men (Cameron et al. 1989; Nasio et al. 1996).
Consider next the studies that examined urethritis. Of these, gonorrhea was the subject of investigation in seven studies. Three concluded that circumcised men were at less risk (Wilson 1947; Parker et al. 1983; Cook, Koutski, and Holmes 1994), and four found no significant difference (Taylor and Rodin 1975; Smith, Greenup, and Takafuji 1987; Donovan, Bassett, and Bodsworth 1994; Laumann, Masi and, Zukerman 1997). Six studies considered non-gonococcal urethritis. Two found circumcised men to be at increased risk (Smith, Greenup, and Takafuji 1987; Newell et al. 1993), and four found no significant difference (Parker et al. 1983; Cook, Koutski, Holmes 1994; Donovan, Bassett, and Bodwsworth 1994; Laumann, Masi, and Zuckerman 1997).
Four of the studies investigated genital warts. Two found circumcised men to be at increased risk (Cook, Koutsky, and Homes 1993; 1994), and two found no difference in the risk faced by circumcised and uncircumcised men (Parker et al. 1983; Donovan, Bassett, and Bodsworth 1994). Van Howe, a well-known opponent of circumcision, judges on the basis of this and other data, that "no solid epidemiological evidence has been found to support the theory that circumcision pre-vents STDs." In his view the "only consistent trend is that uncircumcised males may be more susceptible to GUD, while circumcised men are more prone to urethritis" (Van Howe 1999b, p. 59).
Over the years, the American Academy of Pediatrics has issued a number of position statements about neonatal circumcision. The most recent of these, which was published in 1999 (American Academy of Pediatrics Task Force on Circumcision 1999), had very little to say about STDs. They concluded that "circumcised males may be less at risk for syphilis than are uncircumcised males (p. 691)." Others have been more enthusiastic about the protection circumcision affords against STDs. Schoen, a well-known defender of routine neonatal circumcision, and his colleagues, state that "[s]trong evidence . . . links lack of male circumcision to increased risk for genital ulcer disease, particularly chancroid and syphilis" (Schoen, Wiswell, and Moses 2000, p. 621). Moses, Bailey, and Ronald (1998), in reviewing a host of studies, claim that "there is good concordance for an association between lack of circumcision with chancroid, syphilis, genital herpes, and gonorrhea. Only for urethritis other than gonorrhea and genital warts is the evidence for an effect of circumcision inconclusive (p. 370)."
Human Immunodeefciency Virus
Evaluating the claim that circumcision is associated with a lower incidence of HIV infection15 is greatly facilitated by a number of systematic reviews as well as a meta-analysis. Moses and colleagues reviewed 30 studies (Moses et al. 1994). Twenty-two of these (including two prospective studies) found a statistically significant association between lack of circumcision and HIV infection. (The magnitude of the increased risk for the uncircumcised ranged from 1.5 to 8.4.) Four studies in the review found a trend toward an association, and four studies found no association.
More recently, Weiss, Quigley, and Hayes (2000) performed a meta-analysis of 28 published studies that evaluated the risk factors for susceptibility to HIV-1 infection in men in sub-Saharan Africa. In 21 studies circumcision was associated with a reduced risk of HIV infection, the difference being statistically significant in 14. A higher risk of HIV infection in circumised men was observed in six studies (four from a single area), but none reached statistical significance. Overall, circumcision was associated with a highly significant reduction in the risk of HIV infection (RR 0.52). Significant heterogeneity was observed between the studies, indicating that the magnitude of the protective effect varied in different populations, with the association between circumcision and susceptibility to HIV infection being strongest in high-risk patients.
Van Howe (1999a) reviewed and analysed 33 studies. He concluded that circumcision is associated with an increased risk of acquiring and transmitting HIV. However, Moses, Nagelkerke, and Blanchard (1999) have raised some powerful objections to this study. They have pointed to and explained the methodological flaw of combining raw data for reanalysis, as Van Howe does. That this analysis is flawed stands to reason when one compares the results of the individual studies cited by Van Howe with the conclusions he reaches via his analysis. Of the 33 studies he examined, 16 showed an association between lack of circumcision and in-creased risk of HIV infection, 4 suggested a trend toward this association, 12 demonstrated no association, and only 1 showed an increased risk of HIV among circumcised men. O'Farrell and Egger (2000) have pointed to the same methodological flaws in Van Howe's analysis. Moreover, they have reanalysed the studies he reviewed, and, in contrast to him, they concluded that lack of circumcision is associated with an increased risk of HIV infection. They note that this relationship is only present in groups at high risk for HIV infection.
Reliability of the Primary Data
Numerous objections have been raised against the primary studies on the relationship between circumcision status and sexually transmitted diseases. One problem is that of publication bias - that there is a tendency not to submit or to publish studies that suggest no effect. This is a genuine concern, not only with regard to this issue, but with the whole enterprise of scientific publication. Nevertheless, one is only able to base judgments on available evidence. Other, more important problems, are those of selection bias (how individuals are identified for inclusion in the study), the method of ascertaining circumcision status, the type of study (whether prospective, retrospective, etc.), and the degree to which confounding variables are controlled. While individual studies may be subject to criticism on some or other of these grounds, it is also the case that many primary studies have avoided these pitfalls to varying extents and nonetheless found circumcision associated with lower risk of STDs, including HIV. The re-view by Moses and his colleagues (Moses et al. 1994), for example, included two population-based studies as well as two prospective studies. On the issue of determining a subjectâ€™s circumcision status, Weiss, Quigley, and Hayes (2000) excluded from their analysis those primary studies that used a proxy (such as religion) for circumcision, and many studies have used doctor examination rather than self-report (e.g., Parker et al. 1983; Smith, Greenum, and Takafuji 1987; Cook, Koutsky, and Holmes 1994; Nasio et al. 1996; Tyndal et al. 1996). Many studies controlled for confounding variables. Bailey, Neema, and Othieno (1999) found that even controlling for differences in sexual practices and hygienic behavior, circumcised men were at lower risk. In their meta-analysis, Weiss, Quigley, and Hayes (2000) noted that some adjustment for confounding factors was reported in 15 of 21 studies regarding HIV. Among the factors controlled for were age, ethnic group, marital status, area of residence, sexual behavior, and condom use. Although the con-trolling for confounding variables is much better in the HIV studies, there are some studies on non-HIV STDs that controlled for confounding variables, such as age, ethnicity, marital status, and number of sexual partners (e.g., Cook, Koutsky, and Holmes 1994).
What the above shows, we believe, is that the evidence is stronger and of a better quality for HIV (particularly in high-risk hetereosexual groups) than for other STDs, and that it is stronger for some non-HIV STDs than for others. Although none of the evidence is anywhere near conclusive, it also cannot be ignored. The criticisms of the studies that have been done, although having a certain force, are insufficient to discard those studies and their findings.
There are a number of other issues that are raised in debates about the medical value or risks of circumcision. For instance, circumcision has been claimed to facilitate genital hygiene. This claim is false under some interpretations and true under others. It is not the case that maintaining genital hygiene is very difficult for an uncircumcised man. However, it is the case that it takes slightly more effort than for a circumcised man. That slight difference is not important in itself, but it would be somewhat significant if uncircumcised men, either in the absence of or contrary to any attempts at education about genital hygiene, tended not to exert that extra effort. Unfortunately, there is simply little data, and that uncompelling, on the basis of which any judgment about this matter can be made (Oster 1968; Herzog and Alvarez 1986; Krueger and Osborn 1986; Fergusson, Lawton, and Shannon 1988).
Circumcision has also been said to protect against phimosis16 and paraphimosis.17 Although these conditions can occur only in those who are uncircumcised (or incompletely circumcised), their incidence is very low.18 Thus circumcision, where it is not incompletely done, does prevent these conditions in the small proportion of men who would otherwise have acquired either of them. It has also been claimed that balanitis19 is more common in the uncircumcised. Nevertheless, the incidence remains low (Herzog and Alavarez 1986). Meatitis20 and meatal ulceration occur more often in the circumcised (Mackenzie 1966). There are only a few studies that bear on these issues, and those that do are old. Thus caution is required in the conclusions one draws from this data.
There have been some suggestions that there is a lower incidence of cervical cancer in the female partners of circumcised men. However, there is widespread agreement that there is inadequate data to make such a claim (Niku, Stock, and Kaplan 1995; Moses, Bailey, and Ronald 1998).
Finally, conflicting claims have been made about the relationship between circumcision and sexual pleasure in the man and his female partner. On the one hand, it has been argued that circumcised men experience less sexual pleasure. This has been explained by the keratinization of the exposed glans and the loss of the highly erogenous preputial tissue. However, what little evidence there is on this matter suggests that the circumcised glans is no less sensitive (Masters and Johnson 1966). Moreover, removal of erogenous tissue does not necessarily entail diminished sexual pleasure if sufficient erogenous tissue remains. Others have argued that sexual dysfunction is less common in circumcised men (Laumann, Masi, and Zuckerman 1997) and that the circumcised status is preferred by female partners. Sexual preferences for the circumcised or uncircumcised state will depend on many variables, including culture. It thus seems ill-advised to draw general conclusions from the few studies there have been.
Weighing Up Costs and Benefits
It should be clear from our surveying of the available evidence about circumcision that the practice has both costs and benefits. The most significant cost of neonatal circumcision is the pain that accompanies it. Performing this procedure without adequate analgesia, as is usually the case, is of great moral concern. Given that safe and effective local anesthesia for neonatal circumcision is possible, there is no excuse for failing to use it. Where it is used, this major cost can be eliminated or at least significantly reduced. While circumcision can involve complications, these are mostly minor. Clinically significant negative sequelae are extremely rare. The available evidence suggests that circumcision is protective against the more severe forms of penile cancer and has a small but real effect in reducing the incidence of urinary tract infections. Circumcision is also associated with a lower risk of genital ulcer disease but a slightly in-creased risk of urethritis. At least in high-risk heterosexual groups, circumcision also seems to lower susceptibility to HIV infection.
This would suggest that the potential benefits of neonatal circumcision slightly outweigh the costs, although this is not obviously so. There are a few reasons for being cautious about judging that the balance tips in favor of circumcision. First, the data is incomplete. For instance, the true incidence of serious complications is unknown. Second, not every potential cost and benefit will be equally relevant in every circumstance. For instance, in communities where the incidence of penile cancer and sexually transmitted diseases is very low, the expected benefits of circumcision will be far fewer than in societies in which these conditions are more prevalent. Finally, an overall assessment of the medical costs and benefits of circumcision can-not be made independent of personal value judgments. For example, different people will make different judgments regarding whether reducing the small risk of penile cancer is worth the remote risk of a serious complication from circumcision.
For these reasons, we think that neonatal circumcision cannot unequivocally be said to yield a net medical gain or loss. In other words, it is not something that can be said to be routinely indicated, nor something that is routinely contraindicated. It is a discretionary matter. The decision whether or not to circumcise a child should thus be made by the parents, who, within certain limits, are entitled to employ their own value judgments in furtherance of their child's best interests. These limits are not exceeded in most decisions about neonatal circumcision, given the nature of the medical evidence.
Prior to the last century it was not medical but rather cultural and religious reasons for which circumcision was most often performed. Circumcision continues to be practiced for such reasons by many people. Cultural practices do not have trumping moral weight. That is to say, simply be-cause a practice is culturally valued does not mean that it is morally acceptable. Sometimes a culture treats people in such harmful ways that these peopleâ€™s rights are violated. The practices of widow burning and foot binding are examples. Were it the case that male circumcision unequivocally inflicted as serious harms as do these practices, then its cultural value would be morally overridden. However, the available medical evidence does not support this conclusion and thus such a consideration cannot outweigh the powerful cultural value that circumcision has for many people. Two papers that performed a formal cost-benefit analysis of neonatal male circumcision also reached the conclusion, given the nature of the medical evidence, that cultural and religious considerations should determine whether circumcision is performed (Lawler, Bisonni, and Holtgrave 1991; Ganiats et al. 1991).
This is not to say that people should accept their cultural practices uncritically, even if the weight of evidence does not speak against them. It is all too easy (and common) to privilege those cultural ways to which one is accustomed on account of their familiarity. There is value in stepping back from one's cultural assumptions. When one views male circumcision from another cultural perspective, one can only wonder what possessed ancient people to first think of removing the foreskin. Considered independently, it is about as strange as deciding to remove a part of the earlobe from all children.
This is just the view many people have of clitoridectomy, for example. Of course, in addition to being strange (at least to those outside of the cultures where it is practiced), is also very harmful. It is this harm that separates female genital excision from male circumcision. Nevertheless, the in-ability of many people to step back from their cultural unfamiliarity with genital alterations of young girls is reflected in discussions about how Western societies should relate to female genital excision either in their countries of origin or where immigrants wish to bring the practice to their new Western homes. In such discussions the usual line of argument is that the practice should be entirely eliminated rather than modified to make it less harmful and more akin to male circumcision. More specifically, it is suggested that it would be an un-acceptable strategy to encourage a less damaging form of genital surgery as a way of accommodating cultural ways while minimizing their harmfulness. Excision of the clitoral prepuce is anatomically neither more nor less radical a procedure than removal of the penile foreskin.21 Yet, many of those who would not think twice about circumcising a boy would balk at permitting even the partial removal of a young girl's prepuce.
Some might explain their antipathy only to the latter by arguing that the removal of female genital tissue is historically rooted in misogynistic ideas, because the expressed aim of female genital cutting is often to curb female sexuality. This, the opponents of the practice might say, oppresses women. Thus the fitting response is not to refine the practice but to abolish it. But this explanation reflects the very cultural bias to which we have referred, because it is far from clear that there is an asymmetry between the removal of analogous bits of male and female genital tissue. Consider the following three possibilities:
- If removing the preputial tissue curbs sexuality of both sexes and if, on the basis of curbing female sexuality, the removal of the female prepuce is misogynistic, then the removal of the male foreskin should be viewed as misandristic.
- If removing preputial tissue curbs sexuality but male circumcision is not misandristic because it also affirms the male in the eyes of the community, then female circumcision, which also affirms females in their communities, is not misogynistic.
- Finally, if removing preputial tissue does not curb sexuality in either sex, then the basis for saying that the removal of the clitoral prepuce is misogynistic is eliminated.
The culturally blinded person fails to see that just as female circumcision has been judged, both by its supporters and its opponents, to curb female sexuality, so has male circumcision been said, again by both its supporters and detractors, to curb male sexuality (Maimonides 1956; Gollaher 2000). And they fail to see that just as male circumcision is seen (often simultaneously) as an affirmation of the male, so female circumcision is seen (often simultaneously) as an affirmation of the female. While the removal of clitoris and labia is indeed clearly un-like the removal of the male foreskin, the same cannot be said of the removal of analogous tissues.
There are other differences, of course, between the removal of male and female preputial tissue. For instance, although there is some evidence about the medical value of male circumcision there is none about a comparable benefit in females. There are two possible explanations for this:
- the female procedure has no medical benefits; or
- there might be such benefits but the matter has not received any scientific attention (yet).
However, we suspect that the opposition to excising the clitoral prepuce is based not so much on the absence of medical evidence for a benefit, as on an abhorrence for removing genital tissue from a girl.22 This suggests an asymmetrical judgment about the intrinsic acceptability of removing preputial tissue. This asymmetry can be addressed either by extending the rejection of genital alteration to male circumcision or by withdrawing it from the comparable procedure in females. Until symmetrical judgments are made about comparable procedures, we have every reason to believe that our cultural assumptions are blinding us one way or the other.
We are not endorsing or condemning cultural views in favor of or against (nonharmful) circumcision. We are suggesting that there are cultural biases and that comparable practices (if they really are comparable) should be weighted equally. We are also suggesting that cultural views can themselves be subject to scrutiny and evaluation, and one way this can be done is by reflecting on analogous practices in other cultures to determine whether one's cultural views are consistent.
The Burden of Proof
It may be objected that our argument for the moral acceptability of circumcision rests on a mistaken presumption about when elective surgery is permissible.23 More specifically, it might be said that it is not the mere absence of harm that renders surgery permissible but also the presence of clear and significant net benefit. In other words, it might be argued, it is not sufficient to show that a surgical intervention will not be harmful. There must, in addition, be a demonstrable benefit. However, if we are correct that no clear and significant medical benefits derive from circumcision, there might still be other kinds of benefits.24 Thus the crucial question is whether the relevant presumption should be that
- surgery is impermissible unless it offers clear and significant net medical benefit; or
- surgery is impermissible unless it offers clear and significant net (medical or nonmedical) benefit.
Those who would opt for the first presumption would have to explain why it is that medical benefits are the only relevant ones. It is not as though medical benefits are necessarily or always more important. Some medical benefits are minor and some nonmedical benefits are of great importance. There seems no reason to privilege the one kind of benefit over the other simply because the one is a medical benefit. Nor is it clear why medicine should be used to secure only some kinds of benefits. If education, for example, may be used for medical or cultural benefit and if engineering may be used for social benefit, why may medicine not be used (within appropriate limits) for cultural or other human benefit? Obviously, much more can be said about this issue. However, resolving this issue would take us well beyond a focus on circumcision to a host of other less or uncontested practices, which space constraints prevent.
We have examined both conceptual issues and empirical evidence pertaining to neonatal circumcision of boys. Our conclusion is that circumcision is neither a compelling prophylactic measure nor a form of child abuse. For this reason, nontherapeutic circumcision of infant boys is a suitable matter for parental discretion. In exercising that discretion, religious and cultural factors, though preferably subject to critical evaluation, may reasonably play a role. That our conclusion occupies the popular middle ground between those who condemn the practice outright and those who think it should be routinely performed does not provide grounds for accepting it. The middle way is sometimes the wrong way. In the circumcision debate, however, the evidence and arguments support neither of the extremes.
Note: see also the authors' response to critics of their study.
- American Academy of Pediatrics. 2001. Circumcision: Information for parents. Available from: http://www.aap. org/family/circ.htm. (Accessed 2 Jan. 2001)
- American Academy of Pediatrics Task Force on Circumcision. 1989. Report of the task force on circumcision. Pediatrics 84:388-91.
- —. 1999. Circumcision policy statement. Pediatrics 103(3):686-93.
- Anand, K. J. S., and P. R. Hickey. 1987. Pain and its effects in the human neonate and fetus. New England Journal of Medicine 317(21):1321-29.
- Arnett, R. M., S. Jones, and E. O. Horger III. 1990. Effectiveness of 1% lidocaine dorsal penile nerve block in infant circumcision. American Journal of Obstetrics and Gynecology 163(3):1074-80.
- Bailey, R. C., S. Neema, and R. Othieno. 1999. Sexual behaviors and other HIV risk factors in circumcised and uncircumcised men in Uganda. Journal of Acquired Immune Deficiency Syndromes 22(3):294-301.
- Benatar, D., and M. Benatar. 2001. A Pain in the fetus: Toward ending confusion about fetal pain. Bioethics 15(1): 57-76.
- Benini, F., C. Johnston, D. Faucher, and J. V. Aranda. 1993. Topical anesthesia during circumcision in newborn infants. JAMA 270(7): 850-53.
- Cameron, D. W., J. N. Simonsen, L. J. D'Costa et al. 1989. Female to male transmission of human immunodeficiency virus type 1: Risk factors for seroconversion in men. Lancet ii:403-407.
- Coleman, D. L. 1998. The Seattle compromise: Multicultural sensitivity and Americanization. Duke Law Journal 47:717-83.
- Cook, L. S., L. A. Koutsky, and K. K. Holmes. 1993. Clinical presentation of genital warts among circumcised and uncircumcised heterosexual men attending an urban STD clinic. Genitourinary Medicine 69:262-64.
- —. 1994. Circumcision and sexually transmitted diseases. American Journal of Public Health 84:197-201.
- Dagher, R., M. L. Selzer, and J. Lapides. 1973. Carcinoma of the penis and the anti-circumcision crusade. Journal of Urology 110:79-80.
- Dean, A. L. 1935. Epithelioma of the penis. Journal of Urology 33:252-83.
- Denniston, G. C. 1999. Tyranny of the victims: An analysis of circumcision advocacy. In Male and female circumcision: Medical legal and ethical considerations in pediatric practice, ed. G. C. Denniston, F. M. Hodges, and M. F. Milos, 223-24. New York: Kluwer.
- Denniston, G. C., F. M. Hodges, and M. F. Milos. 1999. Preface. In Male and female circumcision: Medical legal and ethical considerations in pediatric practice, ed. G. C. Denniston, F. M. Hodges, and M. F. Milos, i-vii. New York: Kluwer.
- Derbyshire, S. 1999. Locating the beginnings of pain. Bioethics 13(1):1-31.
- —. 2001. Fetal pain: An infantile debate. Bioethics 15(1):77-84.
- Donovan, B., I. Bassett, and N. J. Bodsworth. 1994. Male circumcision and common sexually transmissible diseases in a developed nation setting. Genitourinary Medicine 70:317-20.
- Fergusson, D. M., J. M. Lawton, and F. T. Shannon. 1988. Neonatal circumcision and penile problems: An eight year longitudinal study. Pediatrics 81:537-41.
- Fetus and Newborn Committee, Canadian Paediatric Society. 1996. Neonatal circumcision revisited. Canadian Medical Association Journal 154:371-72.
- Fleiss, P. M., and F. Hodges. 1996. Neonatal circumcision and penile cancer: Authors' reply. British Medical Journal 313:47.
- Fredman, R. M. 1969. Neonatal circumcision: a general practitioner survey. Medical Journal of Australia 1:117-20.
- Ganiats, T. G., J. B. C. Humphrey, H. L. Taras, and R. M. Kaplan. 1991. Routine neonatal circumcision: a cost-utility analysis. Medical Decision Making 11(4):282-89.
- Gee, W. F., and J. S. Ansell. 1976. Neonatal circumcision: a ten-year overview: With comparison of the Gomco clamp and Plastibell device. Pediatrics 58(6):824-27.
- Ginsberg, C. M., and G. H. McCracken Jr. 1982. Urinary tract infections in young infants. Pediatrics 69:409-11.
- Gollaher, D. 2000. Circumcision. New York: Basic Books.
- Hardner, G. J., T. Bhanalaph, G. P. Murphy, D. J. Albert, and R. H. Moore. 1972. Carcinoma of the penis: Analysis of therapy in 100 consecutive patients. Journal of Urology 108:428-30.
- Herzog, L. W. 1989. Urinary tract infections and circumcision. American Journal of Diseases of Children 143:348-50.
- Herzog, L. W., and S. R. Alvarez. 1986. The frequency of foreskin problems in uncircumcised children. American Journal of Diseases of Children 140:254-56.
- Kirya, C., and M. W. Werthmann. 1978. Neonatal circumcision and penile dorsal nerve blockâ€”A painless procedure. The Journal of Pediatrics 92(6):998-1000.
- Krueger, H., and L. Osborn. 1986. Effects of hygiene among the uncircumcised. The Journal of Family Practice 22(4):353-55.
- Lander, J., B. Brady-Fryer, J. B. Metcalfe, S. Nazarali, and S. Muttitt. 1997. Comparison of ring block, dorsal penile nerve block, and topical anesthesia for neonatal circumcision. JAMA 278(24):2157-62.
- Laumann, E.O., C. M. Masi, and E. W. Zuckerman. 1997. Circumcision in the United States: Prevalence, prophylactic effects and sexual practice. JAMA 277(13):1052-1057.
- Lawler, F. H., R. S. Bisonni, and D. R. Holtgrave. 1991. Circumcision: A decision analysis of its medical value. Family Medicine 23(8):587-93.
- Lenowitz, H., and A. P. Graham. 1946. Carcinoma of the penis. Journal of Urology 56:458-84.
- Littlewood, J. M. 1972. 66 infants with urinary tract infection in first month of life. Archives of Disease in Childhood 47:218-26.
- Mackenzie, A. R. 1966. Meatal ulceration following neonatal circumcision. Obstetrics and Gynecology 28(2):221-3.
- Maden, C., K. J. Sherman, A. M. Beckmann et al. 1993. History of circumcision, medical conditions and sexual activity and risk of penile cancer. Journal of the National Cancer Institute 85:19-24.
- Maimonides, M. 1956. The guide for the perplexed, 2nd ed, trans. M. Friedlander. New York: Dover Publications.
- Malek, R. S., J. R. Goellner, T. F. Smith, M. J. Espy, and M. R. Cupp. 1993. Human papillomavirus infection and intraepithelial, in situ, and invasive carcinoma of penis. Urology 42:159-70.
- Masters, W. H., and V. E. Johnson. 1966. Human Sexual Response. Boston: Little Brown & Company.
- Moses, S., R. C. Bailey, and A. R. Ronald. 1998. Male circumcision: Assessment of health benefits and risks. Sexually Transmitted Infections 74:368-373.
- Moses, S., N. J. D. Nagelkerke, and J. Blanchard. 1999. Analysis of the scientific literature on male circumcision and the risk for HIV infection. International Journal of STD and AIDS 10:626-28.
- Moses, S., F. Plummer, J. E. Bradley et al. 1994. The association between lack of male circumcision and risk for HIV infection: A review of the epidemiological data. Sexualy Transmitted Diseases 21(4):201-10.
- Nasio, J. M., N. J. D. Nagelkerke, A. Mwatha et al. 1996. Genital ulcer disease among STD clinic attenders in Nairobi: Association with HIV-1 and circumcision status. International Journal of STD and AIDS 7:410-14.
- Newell J., K. Senkoro, F. Mosha et al. 1993. A population-based study of syphilis and sexually transmitted disease syndromes in north-western Tanzania. 2. Risk factors and health seeking behaviour. Genitourinary Medicine 69:421-26.
- Niku, S. D., J. A. Stock, and G. W. Kaplan. 1995. Neonatal circumcision. Urologic Clinics of North America 22(1):57-65.
- O'Farrell, N., and M. Egger. 2000. Circumcision in men and the prevention of HIV infection: A "meta-analysis" re-visited. International Journal of STD and AIDS 11(3):137-42.
- Oster, J. 1968. Further fate of the foreskin: incidence of preputial adhesions, phimosis and smegma among Danish schoolboys. Archives of Diseases of Children 43:200-203.
- Parker, J. D. J., and J. E. Banatvala. 1967. Herpes genitalis: Clinical and virological studies. British Journal of Venereal Disease 43:212-16.
- Parker, S. W., A. J. Stewart, M. N. Wren, M. M. Gollow, and J. A. Y. Straton. 1983. Circumcision and sexually transmittable disease. The Medical Journal of Australia 2:288-90.
- Patel, H. 1966. The problem of routine circumcision. Canadian Medical Association Journal 95:576-81.
- Poma, P. A. 1980. Painless neonatal circumcision. International Journal of Gynaecology and Obstetrics 18:308-9.
- Sara, C. A., and C. J. Lowry. 1985. A complication of circumcision and dorsal block of the penis. Anaesthesia and Intensive Care 13(1):79-82.
- Schoen, E. J. 1990. The status of circumcision of newborns. New England Journal of Medicine 322:1308-1311.
- —. 1991. The relationship between circumcision and cancer of the penis. CA Cancer Journal for Physicians 41(5):306-309.
- —. 1996. Neonatal circumcision and penile cancer. British Medical Journal 313:46.
- Schoen, E. J., M. Oehrli, C. J. Colby, and G. Machin. 2000. The highly protective effect of newborn circumcision against invasive penile cancer. Pediatrics 105(3):e36-e39.
- Schoen, E. J., T. E. Wiswell, and S. Moses. 2000. New policy on circumcision - Cause for concern. Pediatrics 105(3): 620â€“23.
- Simonsen, J. N., W. Cameron, M. N. Gakinya et al. 1988. Human immunodeficiency virus infection among men with sexually transmitted diseases: Experience from a center in Africa. New England Journal of Medicine 319(5): 27478.
- Smith, G. L., R. Greenup, and E. Takafuji. 1987. Circumcision as a risk factor for urethritis in racial groups. American Journal of Public Health 77(4):452-54.
- Snellman, L. W., and H. J. Stang. 1995. Prospective evaluation of complications of dorsal penile nerve block for neonatal circumcision. Pediatrics 95(5):705-708.
- Somerville, M. A. 1999. Respect in the context of infant male circumcision: Can ethics and law provide insights? In Male and female circumcision: Medical legal and ethical considerations in pediatric practice, ed. G. C. Denniston, F. M. Hodges, and M. F. Milos, 413-24. New York: Kluwer.
- Speert, H. 1953. Circumcision of the newborn: An appraisal of its present status. Obstetrics and Gynecology 2(2):164-71.
- Spencer, D. M., K. A. Miller, M. Oâ€™Quinn et al. 1992. Dorsal penile nerve block in neonatal circumcision: Chloroprocaine versus lidocaine. American Journal of Perinatology 9(3):214-18.
- Stang, H. J., M. R. Gunnar, L. Snellman, L. M. Condon, and R. Kestenbaum. 1988. Local anesthesia for neonatal circumcision: Effects on distress and cortisol response. JAMA 259(10):1507-11.
- Szmuk, P., T. Ezri, H. Ben Hur et al. 1994. Regional anaesthesia for circumcision in adults: A comparative study. Canadian Journal of Anaesthesia 41:1181-84.
- Taylor, P. K., and P. Rodin. 1975. Herpes genitalis and circumcision. British Journal of Venereal Disease 51:274-277.
- To, T., M. Aghar, P. T. Dick, and W. Feldman. 1998. Cohort study on circumcision of newborn boys and subsequent risk of urinary tract infection. Lancet 352:1813-16.
- Tyndall, M. W., A. R. Ronald, E. Agoki et al. 1996. Increased risk of infection with human immunodeficiency virus type 1 among uncircumcised men presenting with genital ulcer disease in Kenya. Clinical Infectious Diseases 23:449-53.
- Urassa, M., J. Todd, J. T. Boerma, R. Hayes, and R. Isingo. 1997. Male circumcision and susceptibility to HIV infection among men in Tanzania. AIDS 11(1):73-80.
- Van Howe, R. S. 1997. Response to "Determinants of Decision Making for Circumcision" by C. Ciesielski-Carlucci, N. Milliken, and N. H. Cohen (CQ, Vol. 5, No. 2). Cambridge Quarterly of Healthcare Ethics 6:88-89.
- —. 1999a. Circumcision and HIV infection: review of the literature and meta-analysis. International Journal of STD and AIDS 10:8-16.
- —. 1999b. Does circumcision influence sexually transmitted diseases? A literature review. British Journal of Urology International 83(Suppl. 1):52-62.
- Weiss, H. A., M. A. Quigley, and R. J. Hayes. 2000. Male circumcision and risk of HIV infection in sub-Saharan Africa: A systematic review and meta-analysis. AIDS 14(15): 2361â€“70.
- Williamson, P. S., and M. L. Williamson. 1983. Physiologic stress reduction by a local anesthetic during newborn circumcision. Pediatrics 71(1):36-40.
- Wilson, R. A. 1947. Circumcision and venereal disease. Canadian Medical Association Journal 56:54-56.
- Wiswell, T. E., R. W. Enzenauer, J. D. Cornish, and C. T. Hankins. 1987. Declining frequency of circumcision: Implications for changes in the absolute incidence and male to female sex ratio of urinary tract infections in early infancy. Pediatrics 79:338-41.
- Wiswell, T. E., and D. W. Geschke. 1989. Risks from circumcision during the first month of life compared with those for uncircumcised boys. Pediatrics 83(6):1011-15.
- Wiswell, T. E., and W. E. Hachey. 1993. Urinary tract infections and the uncircumcised state: An update. Clinical Pediatrics 32:130-34.
- Wiswell, T. E, and J. D. Roscelli. 1986. Corroborative evidence for the decreased incidence of urinary tract infections in circumcised male infants. Pediatrics 78:96-99.
- Wiswell, T. E., F. R. Smith, and J. W. Bass. 1985. Decreased incidence of urinary tract infections in circumcised male infants. Pediatrics 75:901-903.
- There are some exceptions. See, for example, Somerville 1999.
- Opposition to circumcision of even a consenting adult is a conceivable view, but it is not one we have heard expressed.
- One pair of authors, in speaking about proponents of circumcision remarked: "We sincerely hope that they are more interested in preventing penile cancer than in perpetrating unethical, destructive, mutilative, antisexual, Bronze Age blood rituals on defenceless children" (Fleiss and Hodges 1996). Others referred to an infant's response to circumcision as "screams of agony and protest" (Denniston 1999, 223; emphasis added).
- When an infectious disease is uncommon in a particular population, the individual benefits from widespread vaccination that results in herd immunity. It is not clear, however, that the individual benefits from his or her own vaccination in the context of herd immunity.
- A sucrose nipple and, in Jewish ritual circumcision,a few drops of wine, have been used, but these do not constitute proper analgesia.
- Anand and Hickey's (1987) paper has been widely cited, including by numerous other studies that have investigated the question of neonatal (and late fetal) pain. That neonates can feel pain is accepted by almost everybody. Derbyshire (1999) is an isolated dissenter. We have demonstrated (Benatar 2001) how he misrepresents the scientific consensus, employs a confused and narrow interpretation of the concept of pain, and reaches a conclusion that is not supported by the available evidence. In his attempt to respond to our paper (Derbyshire 2001), he does not engage our criticisms and simply repeats his earlier errors (this time with invective).
- Lander et al. (1997) found that the ring block was more effective than the dorsal penile nerve block.
- Not everybody accepts these statistics. A recent paper cites data that puts the incidence of penile cancer in the United States at 0.6 per 100,000 males and in Denmark at 1 per 100,000 (Schoen et al. 2000, e39).
- There is conflicting evidence about whether self-reportis a reliable way of determining circumcision status. Some studies (e.g., Urassa et al. 1997) suggest it is not, while others (Parker et al. 1983) suggest it is.
- Circumcision beyond infancy has not been as well studied but seems not to confer the same magnitude of benefit.
- Opponents of circumcision typically do not distinguish between the different forms of the disease, and when they do, they do not provide primary data to show that the more severe forms occur in equal or greater degrees in those who are neonatally circumcised.
- Even those papers that conclude that, all things considered, the costs and benefits of circumcision cancel out one another take the prevention of UTI to be a beneficial feature of circumcision. See, for example, Fetus and Newborn Committee, Canadian Paediatric Society (1996); American Academy of Pediatrics Task Force on Circumcision (1999); and American Academy of Pediatrics (2001).
- This is the conclusion of their meta-analysis of nine published studies. Individual studies estimated the increase from 5- to 89-fold.
- This study points to the occurrence of severe complications but overestimates their frequency because it includes a number of cases that were inadequately treated or not treated with antibiotics at all.
- We refer here to susceptibility to HIV infection rather than to infectivity.
- Acquired inability to retract the foreskin.
- Inability of the retracted foreskin to return to its resting position covering the glans.
- See for example, Herzog and Alvarez (1986). Some studies have appropriately cautioned against the overdiagnosis of phimosis, given that it is developmentally normal for the foreskin to be unretractable in young children.
- Inflammation of the glans.
- Inflammation of the meatus.
- At least this is what our discussions with pediatricurologists and gynecologists with interests in this topic suggest.
- One example is the defeat of what has been termed the "Seattle Compromise." The Harborview Medical Center in Seattle was faced with repeated requests from immigrant Somalian mothers to have their daughters circumcised.The mothers indicated that their daughters would be circumcised with or without the doctors. The hospital suggested a compromise procedure whereby the clitoral prepuce would merely be nicked to draw blood. Evidently this would have satisfied at least some of the mothers. However, even thoughno tissuewould have been removedin this proposed procedure, the plan was squashed by those who oppose any nontherapeutic procedure on a girl's genitalia. Those children who would have had the compromise procedure but who were instead subjected to the traditional one were clearly less well off then they would otherwise have been. See Coleman (1998).
- The use of the term surgery is not intended to exclude what are ostensibly surgical procedures but which are per-formed by nondoctors (such as ritual circumcisers).
- The distinction between medical and nonmedical benefits is not as sharp as many people might think, but we ignore this problem here.