376

22 Biotechnology Law Report 376Number 4 (August 2003)© Mary Ann Liebert, Inc.

Social and Ethical Issues in Nanotechnology: Lessons fromBiotechnology and Other High Technologies

JOEL ROTHSTEIN WOLFSON*

INTRODUCTION

BIOTECHNOLOGY CAN BE BROADLY DEFINED as theuse of specially bred living organisms to solve

problems and produce new products or, more nar-rowly, as the intentional alteration of living organ-isms by manipulation of their DNA. In either case,nanotechnology—that is, the creation of molecule-size machines and other devices and the manipula-tion of substances molecule by molecule—will playan increasingly important role. The ability to cus-tom build portions of DNA and other chemicals bitby bit and the potential to create machines that caninteract with, or even penetrate cells of a living or-ganism, will have a profound affect on biotechnol-ogy.

Numerous articles have recently been writtenabout the convergence of biotechnology, nanotech-nology, and other high technologies. For example,in 2002, The National Science Foundation publisheda lengthy set of papers on “Converging Technolo-gies For Improved Human Performance,” which ex-plored numerous aspects of the convergence ofbiotechnology, nanotechnology, information tech-nology, and cognitive sciences. In explaining thebackground for the project, the report begins:

We stand at the threshold of a new renais-sance in science and technology, based on acomprehensive understanding of the structureand behavior of matter from the nanoscale upto the most complex system yet discovered,the human brain. Unification of science basedon unity in nature and its holistic investigationwill lead to technological convergence and amore efficient societal structure for reaching

human goals. In the early decades of thetwenty-first century, concentrated effort canbring together nanotechnology, biotechnol-ogy, information technology, and new tech-nologies based in cognitive science. Withproper attention to ethical issues and societalneeds, the result can be a tremendous im-provement in human abilities, societal out-comes, and quality of life.1

In an article entitled “Nanotechnology 1 Bio-technology 5 Sustainability,” 2 Professor Street ex-plores “the potential for biotechnology and nan-otechnology to become partners for the innovativesolution of technological problems that have beenwith us for some time.” Among the areas high-lighted in the article are advances that might bemade in nanomedicine, biomolecular motors, andthe use of biotechnology/nanotechnology to cleanup the environment, increase food production, andcreate materials such as new plastics and chemicals.Articles on the convergence of biotechnology andnanotechnology have ranged from the popular3 tothe more scholarly.4

*Joel Rothstein Wolfson practices with the firm of Blank RomeComisky & McCauley LLP in Washington, D.C.

1 Converging Technologies for Improved Human Performance,National Science Foundation (2002), p 1 (emphasis added).2 “Nanotechnology1 Biotechnology5 Sustainability,” G. Street,In: Michel J (ed): Proceedings of the Many Facets of InternationalEducation of Engineers. A.A. Balkema Publishers, 2000.3 See, e.g., “Nanotechnology, Biotechnology Come Together,”K. Burns, North County Times, August 19, 2001; “Scientists ofVery Small Draw Disciplines Together,” New York Times C4(Feb. 10, 2003); “Fantastic Voyage: Tiny Pharmacies PropelledThrough the Body Could Result from Cornell Breakthrough inMolecular Motors,” Cornell News (Sept. 7, 1999).4 See, e.g., Merkel RC. Biotechnology as a route to nanotech-nology, Trends Biotechnol 1999;17:271; “New Motifs in DNANanotechnology,” Fifth Foresight Conference on MolecularNanotechnology (1997); West JL, Halas NJ. Applications ofnanotechnology to biotechnology, Curr Opin Biotechnol2000;11:215.

In any event, biotechnologists and public policymakers need to understand the social and ethical is-sues raised by nanotechnology as they impact andmerge with those of biotechnology. This article out-lines some of those issues.

SOCIAL AND ETHICAL ISSUES INNANOTECHNOLOGY

Nanotechnology has an enormous potential to dogood in society. However, like many technologies,its introduction and implementation raise serious so-cietal and ethical issues, both for the scientists whoare developing this technology and for the membersof the public who may benefit from or be exposedto it. The purpose of this paper is to explore someof these societal and ethical issues. The purpose isnot to take policy positions or to suggest solutionsbut merely to raise some of the important social is-sues. In this way, it is hoped that this paper can formthe basis of a discussion on the public policy rami-fications of nanotechnology, from which positionsand solutions can begin to emerge.

Many of the social and ethical issues are the sameas those that affect a wide range of other high tech-nologies. That is, while the technology is new, theissues it raises have been faced before by researchersand society. We need to remind ourselves about thelessons we have already learned about social andethical issues that were raised by biotechnology(such as from regulatory failures in gene therapy),from the development of nuclear technologies, andfrom computer technologies. For those needing abrief introduction to nanotechnology and its poten-

tial benefits to society, a helpful discussion can befound at www.foresight.org/NanoRev/FIFAQ1.html(last visited June 2, 2003).

FIVE VEXING ISSUES

Haves and have-nots of nanotechnology

As we have seen with other technologies, the de-velopment and deployment of nanotechnology willlikely occur first within certain classes of wealthysocieties and then in wealthier nations in general.The effect and challenge of bridging the gaps be-tween classes of haves and have-nots, and thencountries that have and have-not, needs to be con-sidered.

If the nanotechnology gap will be anything likethe gap that exists in ownership of computers andusage of the Internet, the nanotechnology gap be-tween haves and have-nots will pose real societalissues. In 1995, the U.S. Department of Commercepublished its Falling Through the Net: A Survey ofthe ‘Have Nots’ in Rural and Urban America.5 Thereport noted that the gap in the percentages of house-holds with computers between the rich, white, andeducated and poor minorities with less educationwas enormous. A summary of some statistics fromthat survey illustrates the point (Table 1).

This table illustrates the starkness of the divide.In the mid 1990s, those with annual incomes over$75,000 were seven times more likely to own com-

Biotechnology Law Report � Volume 22, Number 4 377

5 Falling Through the Net: A Survey of the ‘Have Nots’ in Ruraland Urban America, U.S. Department of Commerce, July 1995.

TABLE 1. 1995 DEPARTMENT OF COMMERCE REPORT

Percentage of householdsCharacteristic (Urban households) with computers

Income$75,000 or more 64.4$10,000–14,999 9.1$10,000 or less 8.1

RaceWhite 30.3Black 11.8Hispanic 13.2

EducationCollege (4 years or more) 50.7High school 6.1Elementary school 2.8

puters than those with incomes under $15,000. Ur-ban whites had nearly three times the percentageownership of computers than urban blacks had. Peo-ple with college degrees had more than eight timesthe percentage of computers compared with highschool-educated people and more than 17 times thatof people who had only an elementary-school edu-cation.

While the Department of Commerce has reporteda dramatic narrowing of the digital divide since1995, the gap remains large. Comparing the statis-tics from 1995 reveals the effect (Table 2).

Thus, according to the 2000 Report, high-incomehouseholds now had less than four times the per-

centage of computer ownership than householdswith incomes between $10,000 and $15,000. Whitesnow had less than twice the percentage computerownership of blacks. Post-college-educated personshad less than twice the percentage computer own-ership of high school-educated persons.6

Nonetheless, the digital divide in the UnitedStates remains. For example, in this same October2000 report,7 the DEC reported that 77% of house-holds with incomes exceeding $75,000 per yearhad Internet access (60.9% of households with in-comes between $50,000–$75000 had access),whereas only 12.7% of households with incomesof $15,000 or less had access (21.3% of households

378 Biotechnology Law Report � Volume 22, Number 4

6 The DEC 2000 Report noted the achievements:The rapid uptake of new technologies is occurring among

most groups of Americans, regardless of income, education,race or ethnicity, location, age, or gender, suggesting that dig-ital inclusion is a realizable goal. Groups that have tradition-ally been digital “have nots” are now making dramatic gains:

� The gap between households in rural areas and householdsnationwide that access the Internet has narrowed from 4.0percentage points in 1998 to 2.6 percentage points in 2000.Rural households moved closer to the nationwide Internetpenetration rate of 41.5%. In rural areas this year, 38.9% ofthe households had Internet access, a 75% increase from22.2% in December 1998.

� Americans at every income level are connecting at far higherrates from their homes, particularly at the middle income lev-els. Internet access among households earning $35,000 to$49,000 rose from 29.0% in December 1998 to 46.1% in Au-gust 2000. Today, more than two-thirds of all householdsearning more than $50,000 have Internet connections (60.9%for households earning $50,000 to $74,999 and 77.7% forhouseholds earning above $75,000).

� Access to the Internet is also expanding across every educa-

tion level, particularly for those with some high school or col-lege education. Households headed by someone with “somecollege experience” showed the greatest expansion in Inter-net penetration of all education levels, rising from 30.2% inDecember 1998 to 49.0% in August 2000.

� Blacks and Hispanics still lag behind other groups but haveshown impressive gains in Internet access. Black householdsare now more than twice as likely to have home access thanthey were 20 months ago, rising from 11.2% to 23.5%. His-panic households have also experienced a tremendous growthrate during this period, rising from 12.6% to 23.6%.

� The disparity in Internet usage between men and women haslargely disappeared. In December 1998, 34.2% of men and31.4% of women were using the Internet. By August 2000,44.6% of men and 44.2% of women were Internet users.

� Individuals 50 years of age and older—while still less likelythan younger Americans to use the Internet—experienced thehighest rates of growth in Internet usage of all age groups:53% from December 1998 to August 2000, compared to a35% growth rate for individual Internet usage nationwide.

7 Falling Through the Net: Toward Digital Inclusion: A Reporton Americans’ Access to Technology Tools, U.S. Departmentof Commerce, October 2000.

TABLE 2. 2000 DEPARTMENT OF COMMERCE REPORT

Percentage of householdsCharacteristic (Urban households) with computers

Income$75,000 or more 86.2$10,000–14,999 22.4$10,000 or less 15.1 ($5,000–$9,999)

23.6 (below $5,000)Race

White 57.3Black 33.3Hispanic 34.2

EducationCollege (4 years or more) 75.3High school 38.3Elementary school 13.7

with incomes between $15,000 and $25,000 hadaccess).8

The gap in Internet usage between United Statesand Europe on one hand, and the rest of the worldon the other, remains stark. The BBC recently re-ported that, “Black and Hispanic households are ap-proximately one-third as likely to have home Inter-net access as households of Asian/Pacific Islanderdescent, and roughly two-fifths as likely as whitehouseholds.” Internationally, it noted “more thanhalf of Internet users are from the USA despite mak-ing up just 4.7% of the total world population.” TheBBC report further noted that North America has57% of Internet Users, Europe 21.7%, Asia 17%,South America 3%, the Middle East 0.5%, andAfrica 0.8%.9

If nanotechnology will really offer the promise ofdramatic increases in length of life or the ability to

clean up toxic or household pollutants or to manu-facture superior goods with very low expendituresof energy and even without access to expensive rawmaterials, it will offer these benefits only to thosewho have access to the technology. If the lessons ofthe past with computers and the Internet are anygauge, the gap between rich and poor nations andclasses, and between developed and developing na-tions, will be dramatic.

Similarly, as we have learned from the progressmade in life expectancy from the right combinationof AIDS drugs, technological advances can quicklyhave important positive effects on the quality andlength of life but only to those to whom the tech-nology is available. The high cost of AIDS drugshas meant that longer life is not available to thehave-nots either within American society or in de-veloping countries.10 The situation was summed up

Biotechnology Law Report � Volume 22, Number 4 379

8 The 2000 Report summarized the depth of the divide:

Nonetheless, a digital divide remains or has expandedslightly in some cases, even while Internet access and computerownership are rising rapidly for almost all groups. For exam-ple, our most recent data show that divides still exist betweenthose with different levels of income and education, differentracial and ethnic groups, old and young, single and dual-parentfamilies, and those with and without disabilities.

� People with a disability are only half as likely to have accessto the Internet as those without a disability: 21.6% comparedto 42.1%. And while just under 25% of people without a dis-ability have never used a personal computer, close to 60% ofpeople with a disability fall into that category.

� Among people with a disability, those who have impaired vi-sion and problems with manual dexterity have even lowerrates of Internet access and are less likely to use a computerregularly than people with hearing difficulties. This differ-ence holds in the aggregate, as well as across age groups.

� Large gaps also remain regarding Internet penetration ratesamong households of different races and ethnic origins. AsianAmericans and Pacific Islanders have maintained the highestlevel of home Internet access at 56.8%. Blacks and Hispan-ics, at the other end of the spectrum, continue to experiencethe lowest household Internet penetration rates at 23.5% and23.6%, respectively.

—Large gaps for Blacks and Hispanics remain when mea-sured against the national average Internet penetration rate.—The divide between Internet access rates for Black householdsand the national average rate was 18 percentage points in Au-gust 2000 (a 23.5% penetration rate for Black households, com-pared to 41.5% for households nationally). That gap is 3 per-centage points wider than the 15 percentage point gap thatexisted in December 1998.

—The Internet divide between Hispanic households and thenational average rate was 18 percentage points in August 2000(a 23.6% penetration rate for Hispanic households, compared

to 41.5% for households nationally). That gap is 4 percentagepoints wider than the 14 percentage point gap that existed inDecember 1998.

—With respect to individuals, while about a third of the U.S.population uses the Internet at home, only 16.1% of Hispanicsand 18.9% of Blacks use the Internet at home.

—Differences in income and education do not fully accountfor this facet of the digital divide. Estimates of what Internetaccess rates for Black and Hispanic households would havebeen if they had incomes and education levels as high as thenation as a whole show that these two factors account for aboutone-half of the differences.� With regard to computer ownership, the divide appears to

have stabilized, although it remains large.—The August 2000 divide between Black households and

the national average rate with regard to computer ownershipwas 18 percentage points (a 32.6% penetration rate for Blackhouseholds, compared to 51.0% for households nationally).That gap is statistically no different from the gap that existedin December 1998.

—Similarly, the 17 percentage point difference between theshare of Hispanic households with a computer (33.7%) and thenational average (51.%) did not register a statistically signifi-cant change from the December 1998 computer divide.

—Individuals 50 years of age and older are among the leastlikely to be Internet users. The Internet use rate for this groupwas only 29.6% in 2000. However, individuals in this age groupwere almost three times as likely to be Internet users if theywere in the labor force than if they were not.9 “Plugging into the revolution,” Jane Black, BBC Online, Oc-tober 14, 1999, http://news.bbc.co.uk/1/hi/special_report/1999/10/99/information_rich_information_poor/467899.stm.(Last visited June 2, 2003).10 See, e.g., “Death Watch: The Global Response to AIDS inAfrica World Shunned Signs of the Coming Plague,” BartonGellman, Washington Post, July 5, 2000; Page A1; “The Endof AIDS? The plague continues, especially for the uninsured,but new drugs offer hope for living with HIV,” John Leland,Newsweek December 2, 1996.

this way by the Journal of the American MedicalAssociation11:

Of the more than 21 million adults estimatedto be living with HIV, about 90% live in the de-veloping world, where economic conditionsmake it extremely unlikely they will benefitfrom the expensive new antiretrovirals that haveproven so effective in managing the disease.

“Many, if not most, lack any access to even ba-sic pain-relieving drugs or treatment for their op-portunistic infections,” said Peter Piot, MD, execu-tive director of the Joint United Nations Program onHIV/AIDS.

While combination therapies can run as high as$18,000 a year, most African countries can affordto spend less than $10 a day on one person’s healthcare, said James McIntyre, MD, an obstetrician-gy-necologist at Baragwanath Hospital in Soweto,South Africa.

Speaking at the 11th International AIDS Confer-ence in Vancouver, Jonathan M. Mann, MD, MPH,professor of epidemiology and international healthat Harvard School of Public Health, said the ad-vances in HIV care illustrate the widening chasmbetween rich and poor nations:

“The injustice is stark: Drugs are avail-able—at best—to the less than 10% of theworld’s people with HIV/AIDS in the indus-trialized world . . .

“And even though medical care inequitiesalways have been ‘the tragic rule’ separatingthe haves from the have-nots, the HIV/AIDSpandemic is ‘profoundly different’ in one way. . .

For in AIDS, we all started in the sameplace—with the same lack of treatment andwith the same hopes—and the unfairness hasarisen right before our eyes.”

But there is hope. An initiative has been launchedto get AIDS drugs to poor countries at low cost. Itwas summarized this way in a recent Time article:

. . . increasingly, poor countries and AIDSadvocates are finding ways to shift the balance.. . . So a virtually identical version of the an-tiretroviral combination cocktail that sells for$10,000 to $15,000 a year in the U.S. costs

$3,000 in Brazil and less than $1,000 in India.And when Brazil decided to provide the ge-neric drugs free to all its AIDS victims, it dis-proved the argument that poor countriescouldn’t master the complex regime of AIDSpills. The government set up effective clinics,and reports indicate that Brazilian patients taketheir medicine as meticulously as AmericanAIDS sufferers do . . .

. . . .For five years, unAIDS (the Joint United

Nations program on HIV/AIDS) jawboned thecompanies to set lower prices for developingcountries. Finally, just before the internationalAIDS conference held last July in Durban,South Africa, five major pharmaceuticalsjoined an “Accelerated Access” program to ne-gotiate 60% to 80% reductions in AIDS-drugprices for poor nations.12

Similarly, the initiative to wire every school inthe United States with Internet access shows that so-ciety can narrow the technology gap if it sees theproblem and confronts it directly:

In response to the educational opportunitiesmade available by dramatic technological in-novations in the early and mid-1990s, U.S.Secretary of Education released the nation’sfirst educational technology plan in 1996, Get-ting America’s Students Ready for the 21stCentury: Meeting the Technology LiteracyChallenge. This plan presented a far-reachingvision for the effective use of technology in el-ementary and secondary education to help thenext generation of school children to be bettereducated and better prepared for the evolvingdemands of the new American economy.

Given that many schools and classroomshave only recently gained access to technologyfor teaching and learning, the positive out-comes of these studies suggest a future for ed-ucation that could be quite bright if the nationmaintains its commitment to harnessing tech-nology for education.

380 Biotechnology Law Report � Volume 22, Number 4

11 “Special Reports: New Drugs Have Limited Impact Glob-ally,” JAMA HIV-AIDS Information Center, 1999, www.ama-assn.org/special/hiv/newsline/ special/amnews/amn0916a.htm.12 “Paying for AIDS Cocktails: Who Should Pick up the Tabfor the Third World?” Time, Feb. 12, 2001.

The adoption of new and emerging tech-nologies by schools and classrooms offers evenmore reason to be hopeful. With sufficient ac-cess and support, teachers will be better ableto help their students comprehend difficult-to-understand concepts and engage in learning,provide their students with access to informa-tion and resources, and better meet their stu-dents’ individual needs. If we take advantageof the opportunities presented to us, technol-ogy will enhance learning and improve studentachievement for all students.

Working together to achieve these goalsconstitutes a major leadership imperative fac-ing those seeking widespread improvements inteaching and learning. As a nation, we shouldpledge to meet these new goals.13

Nanotechnology as a terrorist weapon

Because of its microscopic size, easy dispersal,self-replication, and potential to inflict massiveharm on persons, machines, or the environment,nanotechnology makes a tempting terrorist weapon.

Since September 11, 2001, concerns about theconversion of useful machines into terrorist weaponshas been heightened. If rogue states and groups canacquire biological and chemical weapons of massdestruction, then surely they can learn to use nan-otechnology.

The dangers of nanotechnology as a terroristweapon are easy to see. First, a nano-robot that canoperate within a human body could easily be pro-grammed to destroy rather than heal. Because oftheir small size, and because they might be quicklyredesigned to avoid the latest counter-measure,nano-machines pose a potent threat as a terroristweapon of the future. Similarly, nano-machinescould be designed to attack machines, rather thanhumans. They could be made to destroy defensiveweapons, bring electrical generators to a halt, or eataway at protective encasements or linings arounddangerous environments. Moreover, because oftheir small size, they might be easily dispersed inthe air or through the water. Their transport couldbe hard to detect. Finally, if these nano-machinesare programmed to be self-replicating or self-repli-cating and mutating, the danger they pose could bevery hard to contain.

Nanotechnology might someday permit one to as-semble, molecule by molecule or chain by chain,

any compound one desires. Thus, terrorists coulduse nano-machines to assemble pure mixtures ofdangerous toxins, even if they have no access to theunderlying living creature that normally creates thattoxin or to the raw material needed to produce thetoxin. That is, at least theoretically, a nano-machinecould build the anthrax toxin, molecule by moleculeor at least chain by chain, in great abundance, evenif the terrorist had no access to the spore-formingbacterium Bacillus anthracis. The recent success atassembling the polio virus, without even the aid ofnanotechnology, well illustrates this threat,14 asdoes the initiation of a project to remove the genesfrom Mycoplasma gentilatium and replace themwith a pared-down and artificially constructed stringof DNA with just enough genetic material to get thecell going again.15 Terrorists could take relativelyinnocuous forms of a toxin or chemical and, by mak-ing a small addition to or deletion from the naturalstructure, change it into one far more dangerous.

On the other hand, as has been highlighted in therecent debate about the nuclear capabilities of Iraq,there may be choke points in the development ordeployment of weapons that prevent their use byrogue groups. For example, the United States hasalleged that Iraqi scientists may largely possess theknow-how to build a nuclear weapon but have beenunable to gain access to sufficient quantities ofweapons-grade enriched uranium.16 Although it isnot clear that these kinds of non-proliferation strate-gies and choke points exist in the exploitation ofnanotechnology, the quick identification of thesechoke points and the rapid development of a globalconsensus on implementing non-proliferation strate-gies would be crucial. If such strategies do not ex-

Biotechnology Law Report � Volume 22, Number 4 381

13 See, e.g., e-Learning: Putting a World Class Education atthe Fingertips of our Children, U.S. Department of Education,December 2002.14 See, e.g., “Mail-Order Molecules Brew a Terrorism DebateVirus Created in Lab Raises Questions of Scrutiny for DNASuppliers,” Rick Weiss Washington Post, Wednesday, July 17,2002, A01.15 See, e.g., “Creating Living Things,” Editorial, WashingtonPost, November 23, 2002, A22; “Nothing Wrong with a LittleFrankenstein,” Chris Mooney, Washington Post, December 1,2002, B01.16 See, e.g., “Blair: Iraq Can Deploy Quickly; Report PresentsNew Details on Banned Arms,” Glenn Frankel WashingtonPost, September 25, 2002, A01; “Observers: Evidence for WarLacking; Report against Iraq Holds Little That’s New,” DanaPriest and Joby Warrick, Washington Post, September 13, 2002,A30.

ist, then policy makers need to begin to develop ro-bust defense mechanisms to deal with the potentialnanotechnology terrorist threat.

Inadvertent release or inadvertent spread ofnanotechnology

As we have learned with other technologies, sci-entists had thought they had proven methods to pre-vent the inadvertent spread of biotechnology intothe wider environment. They were wrong. Nan-otechnologists face these same risks.

Well-intentioned and expert bioengineering sci-entists were confident that genetically engineeredplant seeds would not be able to migrate into non-engineered fields and would not enter the humanfood chain by accident. They were wrong. Geneti-cally altered seeds and products have been discov-ered in human foods (such as taco shells), and seedsintended for animal feed were planted by farmersand spread into fields of non-engineered crops.17

Similarly, kernels from an experimental corn plantaltered to produce a pharmaceutical product mayhave contaminated a subsequent soybean crop in-tended for human consumption.18

Likewise, food experts were confident that theycould control or exclude the disease agent thatcauses “mad cow” disease from human food chains.They, too, were wrong. That failure has led to massanimal kills causing enormous social and economiccosts for farmers and society and has had wide-rang-ing effects, including an erosion of public confi-dence in government and in the current means toensure safe food supplies.19

Nanotechnologists argue that inadvertent spreadwill not happen because nano-machines need a con-fined source of power, like a battery. They arguethat any inadvertent release is not likely to have sig-nificant detrimental effects, because the nano-ma-chines will simply run out of energy quickly. Thisassumption may be naïve. Scientists have alreadypostulated that nano-machines could be built to relyon energy sources from the environment aroundthem. Moreover, as lovers of electronic gadgetsknow, batteries are becoming better, and power re-quirements are lessening. As a result, while a nano-machine may eventually fail for lack of power, mil-lions of them, inadvertently released, could do greatdamage before that eventuality came true.

It is important to keep in mind that the risk of theinadvertent spread of nanotechnology is less of a

concern in the near term because most nanotech-nology is in the early experimental or developmen-tal stage. Just as scientists have been working withdeadly pathogens in laboratories across the worldfor a long time and have established effective pro-tocols that protect researchers and the general pub-lic from the inadvertent escape of these pathogensfrom facilities that study or genetically alter them,research protocols should be able to protect the pub-lic from an inadvertent spread or release of nan-otechnology during the developmental stages. TheForesight Guidelines on Molecular Nanotechnologyis one attempt to establish principles to guard againstthe inadvertent release of nanotechnology.20

Nonetheless, inadvertent release or spread of nan-otechnology during deployment remains a sseriousrisk. Scientists and policy makers need to keep therisk in mind and devise coordinated contingencyplans to deal with the eventuality.

Who, if anyone, should regulate nanotechnology?

As with any new technology, the question ofwhether there should be regulation of nanotechnol-ogy is an important one that needs to be resolvedearly in its lifecycle. First, one must ask whethernanotechnology should be subject to comprehensiveor more limited subject matter regulation or be leftlargely unregulated. Second, one must ask if thelevel of regulation should be different depending onwhether the activity at issue is research and devel-opment or commercial deployment. Third, becausecertain types of nanotechnology are likely alreadysubject to various decentralized regulatory regimes,a subsidiary question is whether regulation of nan-otechnology should be centralized in one agency or

382 Biotechnology Law Report � Volume 22, Number 4

17 See, e.g., “Gene-Altered Canola Can Spread to NearbyFields, Risking Lawsuits,” Jill Carroll, Wall Street Journal, June28, 2002, B6.18 See, e.g., “ProdiGene-Modified Corn Plant Nearly Gets intoU.S. Food Supply,” Scott Kilman, Wall Street Journal No-vember 13, 2002.19 See, e.g., “In Europe, a Unity of Distrust,” Jim Hoagland,Washington Post, February 1, 2001, A21; “Japan to Test 1 Mil-lion Cattle for ‘Mad Cow’; Concerns Grow after First CaseBotched,” Kathryn Tolbert, Washington Post, September 20,2001, A30; “Beef’s Battles in the Midst of a Comeback; RedMeat Faces Another Image Crisis,” Douglas Hanks III, Wash-ington Post, March 28, 2001, F1.20 Foresight Guidelines on Molecular Nanotechnology, origi-nal version 1.0, February 21, 1999; revised draft 3.7, June 4,2000, Foresight Institute and Institute for Molecular Manufac-turing.

continue to be decentralized. Finally, there is thequestion of whether self-regulation, governmentalregulation, or a mixture of the two is the best ap-proach.

What level of regulation?

The question of what level of regulation to im-pose on an emerging technology is complex. Arange of industries (technological and non-techno-logical) are subject to comprehensive state or fed-eral regulatory schemes in the United States. Ex-amples are the utilities, nuclear power, foods, drugsand cosmetics, and securities. Many industries, suchas the sale of cars, are unregulated in large part, butcertain activities or practices are the subject of tar-geted regulations or prohibitions. Other industries,such as the Internet, software, and consumer appli-ances, are largely unregulated. Analogies and dis-tinctions can be drawn between nanotechnology andindustries in each of these categories, as these threecategories cover the range of possible regulatory ac-tivity, one category must be chosen.

There are two basic motivations for comprehen-sive regulatory schemes—natural monopoly andoverriding public harm. Utilities, such as electric-ity, gas, and telephone service, as well as activitiessuch as broadcasting, have been subject to compre-hensive regulation because they are important ser-vices that form natural monopolies so that there areinadequate free market forces to control quality, ac-cess, prices, and terms of service.21 Nanotechnol-ogy does not form a natural monopoly.

On the other hand, nuclear power, drugs, cos-metics, and foods have been subject to comprehen-sive regulation, not because of their monopoly na-ture, but because of the danger they pose to humansfrom poor quality, poor risk management, falseclaims, or inadequate testing. Even where physicalinjury is not present but harm can result from poorquality, poor risk management, and false claims,comprehensive regulation has been imposed, suchas in the stock market and securities area. As high-lighted elsewhere in this paper, nanotechnology canpose a significant risk to humans, like the risksposed by these regulated industries. Whether the riskis similar enough to warrant such comprehensiveregulation is the open question. I note, however, thatthe regulatory concern about nanotechnology differsfrom that of some of the industries listed above inthat it does not stem primarily from a fear of harm

to the public from false claims about the product butmore from poor quality or poor risk management.Thus, nanotechnology is like, and yet not like, com-prehensively regulated industries.

Nanotechnology is also both like and not likesome focused-regulation industries. Like health careproviders or used car sales, poor nanotechnologypractices may not necessarily lead to an immediatethreat to human life, but their potential for health oreconomic loss to consumers is great enough to havemotivated legislatures to impose detailed regula-tions in certain areas (such as disclosure of interestrates on loans or protection of privacy of health in-formation) or to prohibit certain kinds of undesir-able practices (such as outlawing kickbacks or en-acting lemon laws). There are likely to be particularconcerns about nanotechnology or its uses thatmight prompt legislatures to enact focused legisla-tion to curb or regulate particular parts of the in-dustry.

Finally, nanotechnology is also both like and un-like unregulated industries. Even where defectiveproducts can cause injury and death, such as withheavy equipment or consumer appliances, mostproducts are not regulated. Rather, the legislaturesleave to tort law (such as personal injury lawsuitsor antitrust actions) the job of “regulating” bad con-duct. Also, one can make the argument that the samemarket forces that shape and restrain bad practicesin unregulated industries and the same possibility ofrapid innovation from a lack of regulation that hashelped fuel the rapid growth of the Internet and thesoftware industries are needed to propel the devel-opment of nanotechnology.

Developmental-versus deployment-stageregulation

Sometimes, a level of regulation for develop-mental activities is either not imposed or is imposed

Biotechnology Law Report � Volume 22, Number 4 383

21 Although even this view is subject to challenge and debate.The deregulation of many telecommunications services throughFCC action before and since the Federal TelecommunicationsReform Act of 1996, Pub. LA. No. 104-104, 110 Stat. 56 (1996)highlights the fact that the conclusion that utilities from naturalmonopolies has been subject to rethinking. On the other hand,local telephone services remain regulated, and commentatorsdisagree about whether telecommunications deregulation was agood thing. See, e.g., “How The Bells Stole America’s DigitalFuture: A NetAction White Paper,” Bruce Kushnick, 2001,reprinted at http://www.netaction.org/broadband/bells/ (last vis-ited June 2, 2003).

at a different level than when the same activity isused later commercially or more generally. Psycho-logical counseling techniques are an example. Whena researcher into family dynamics or psychologywishes to perform human experimentation (withfederal funding22), relatively strict regulation is im-posed.23 However, when the same techniques areactually put into practice by school counselors orministers, there is little or no regulation. On theother hand, laboratory development of new foodsand cosmetics is relatively unregulated. However,their sale (deployment) to the public is subject torelatively comprehensive regulation, and for someactivities, because of the dangers during experi-mentation, deployment, and even disposal, such aswith nuclear materials, comprehensive regulation isimposed from cradle to grave.

Policy makers must decide which model appliesto nanotechnology. The answer may depend on theuse. Nanotechnology regulation is not being writtenon a blank slate: there are a host of existing regu-latory schemes that will affect its development anddeployment. This fact leads us into the next topic,“Is the existing decentralized approach to nan-otechnology regulation adequate?”

Centralized versus decentralized regulation

One aspect of the debate about which category—comprehensive regulation, focused regulation, orunregulated—should be used with nanotechnologyis whether existing decentralized regulatory schemesare already focused on the major risks so that nofurther action should be taken at this time. It is likelythat some nanotechnology will be used to createmedical devices subject to the control of the Foodand Drug Administration. It is likely that some nan-otechnology will be used to deploy products regu-lated by the Environmental Protection Agency. Nan-otechnology used in human medical research willbe naturally subject to regulation by the Departmentof Health and Human Services and hospital reviewboards.24 The question presented is whether the ex-isting decentralized approach is adequate.

Under federal law, medical devices25 are subjectto FDA jurisdiction under the Food, Drug, and Cos-metic Act (FDCA) to ensure that they are “safe andeffective.” “Safe” means that the probable benefitsto health in its intended use outweigh any probablerisks of harm or injury by the device. “Effective”means that the device does what it is supposed to

do in a reliable fashion. Under the Safe Medical De-vices Act of 1990 and the Medical Device Amend-ments of 1992, the FDA was granted greater post-market controls, such as user reporting of device-related deaths or serious injuries to provide an earlywarning system for device complications or failures.Nanotechnology robots that are introduced into ahuman body to repair it would seem to be medicaldevices under the FDCA.

Foods and food additives are also regulated bythe FDA under the FDCA, which bans the intro-duction or delivery into interstate commerce of“misbranded” or “adulterated” food.26 “Misbrand-ing” is the use of misleading labeling and packag-ing, as well as false representations as to quality.27

Section 402 of the FDCA defines “adulteration” asthe addition of poisonous or deleterious substancesto food.28 There are General Standards for adulter-ation,29 and if necessary, the FDA can prescribeSpecial Standards for particular types of adulter-ations.30 Nanotechnology that either creates food orfood additives, or the misbranding of nanotechnol-ogy that is used in the human food supply wouldappear to be governed by existing FDA law.

384 Biotechnology Law Report � Volume 22, Number 4

22 See discussion of federal regulation of human research ac-tivities elsewhere in this document.23 This topic is discussed in more detail elsewhere in this arti-cle.24 This decentralized approach was reinforced in 1986, whenthe federal government completed the “Coordinated Frameworkfor Regulation of Biotechnology,” 51 F.R. 23,302–23,350(1986), which has been characterized as “establish[ing] the pol-icy that a product of biotechnology should be regulated ac-cording to its composition and intended use, rather than by themethod used to produce it.” The Regulation of Biotechnology,Randy Vines, Virginia Tech Publication Number 443-006, May2002.25 Federal Food, Drug, and Cosmetic Act, §201(h) defines “de-vice” as “an instrument, apparatus, implement, machine, con-trivance, implant, in vitro reagent, or other similar or relatedarticle, including any component, part, or accessory, which is. . . recognized in the official National Formulatry, or the UnitedStates Pharmacopoeia, or any supplement to them, intended foruse in the diagnosis of disease or other conditions, or in thecure, mitigation, treatment, or prevention of disease, in man orother animals, or intended to affect the structure or any func-tion of the body of man or other animals, and which does notachieve its primary intended purposes through chemical actionwithin or on the body of man or other animals and which is notdependent upon being metabolized for the achievement of itsprimary intended purposes.”26 21 USC §331.27 21 USC §342.28 21 USC §342.29 21 USC §342(a).30 21 USC §342(a)(2).

Under Section 505 of the FDCA and 351 of thePublic Health Service Act,31 drugs must be subjectto premarket approval for their labeled uses. The in-troduction of a misbranded drug is prohibited.32

Again, nanotechnology-created drugs appear to becovered under current FDA regulation.

Pesticides are regulated under the Federal Insec-ticide, Fungicide and Rodenticide Act (FIFRA) of1947.33 Pesticides cannot be sold unless they areregistered and properly labeled under the Act.34 Ap-plicators who use a pesticide unlawfully are subjectto written warning, citation, and fines from theEPA.35 Although it is not clear, nano-machines andcertainly nanotechnology-created pesticides may beunder the current jurisdiction of the EPA.

A significant role in the above regulatory schemeis given to the United States Department of Agri-culture’s Animal and Plant Health Inspection Ser-vice (APHIS). Under the Plant Protection Act of2000,36 the Secretary of Agriculture has the powerto prohibit or restrict imports, exports, or interstatemovements of plants, plant pests, noxious weeds,and biological control organisms. Similarly, the An-imal Health Protection Act (AHPA) of 200237 con-solidates various powers of the USDA in the areaof regulation of animals and technologies that affectanimals and permits the Secretary to prohibit or re-strict entry of any animal or related material if nec-essary to prevent spread of any livestock pest or dis-ease. The Secretary may also prohibit or restrictexports if necessary to prevent the spread of live-stock pests or diseases from or within the U.S.

Regulation of research on human subjects that is“conducted, supported or otherwise subject to reg-ulation by any Federal Department or Agency” isoverseen by the Office for Human Research Pro-tections. 45 CFR Part 46 details the kinds of con-trols that are imposed on human research studies,including Institutional Review Boards and informedconsent procedures.

The above are just some examples of how the reg-ulation of nanotechnology would likely evolve if nodebate is initiated into the benefits and risks of a de-centralized versus a centralized regulatory regime.On one hand, if nanotechnology medical devices ornanotechnology-created pesticides are more likecurrent medical devices and current chemical pesti-cides than they are like other nanotechnologies andnano-machines, then decentralized regulation makessense. On the other hand, if nanotechnology expertsare few and expensive, or nanotechnology risks are

more alike across a range of applications, then itmay be better to concentrate oversight within a sin-gle agency so that it is able to view the entire scopeof use of nanotechnology and form “nano-centric”regulatory programs. A recent article in the Wash-ington Post (commenting on a report by the PewInitiative on Food and Biotechnology) notes, in re-lation to genetically altered fish, for example, thatwhile the FDA has jurisdiction to regulate the foodhazards that might be created by such fish, it has nopower to investigate the environmental hazards re-lated to an accidental release of the fish; that is, nocentral federal agency has the power to viewbiotechnology (or nanotechnology) as a whole.38

The point here is that unless a debate on the meritsof decentralized regulation of nanotechnology be-gins sooner rather than later, agencies will naturallybegin to regulate the nanotechnology within theirexisting reaches. Such agencies will then have avested interest in a decentralized approach to nan-otechnology regulation. Such ingrained interestsmay make a change to central regulation more dif-ficult to achieve (assuming the right answer is cen-tralized regulation).39

Governmental or self regulation?

Finally, one must ask whether regulation shouldbe governmental or self or a mixture thereof. Manyregulatory schemes rely primarily on a series of gov-ernmentally created regulations and government en-forcement programs. Examples are food and drugregulation and environmental regulation. Agenciesimpose a series of operational, record keeping, andreporting duties on the industries or activities within

Biotechnology Law Report � Volume 22, Number 4 385

31 Section 505 of the Food, Drug, and Cosmetic Act, 21 USC355(d); §351 of the PHSA, 42 USC §262.32 21 USC §331–334.33 7 USC §§135–136y.34 7 USC §136a.35 7 USC §136i, 136j-l.36 7 USC §§7711–7758.37 Part of the Farm Bill of 2002.38 “Old Laws, New Fish: Environmental Regulation of Gene-Altered Foods is a Gray Area,” Justin Gillis, Washington Post,January 15, 2003, E01.39 Another issue that must be balanced is the extent to whichparticular regulatory agencies have expertise that must bebrought to bear versus the amount to which they are influencedby the industry they are supposed to regulate. In a similar vein,some agencies are seen as weak regulators, given their statu-tory charters or present leadership, and others are seen as toozealous or ideological.

their domain. While private industry may suggestregulations or comment on the advisability of pro-posed regulations, the governmental agency ulti-mately has the final say. Moreover, violations of theregulations are prosecuted primarily by govern-mental agencies or prosecutors.

On the other hand, a self-regulatory scheme, asin the securities area, leaves the creation of rulesand their primary enforcement to the industry mem-bers themselves. In the securities industry, the Se-curities and Exchange Act of 1934 (SECA) providesthat no broker or dealer (with minor exceptions) mayperform transactions in securities unless it is a mem-ber of a “Securities Association.”40 The SECA thenrequires that Securities Associations be registeredwith the U.S. Securities and Exchange Commission(SEC).41 A Securities Association must fill out avery detailed application that meets a long series oftests set out in the statute before it can be approvedby the SEC. Among these are requirements that gov-ernance of the association be representative of themembers, that the rules provide for a code of con-duct that meets certain standards, and that the asso-ciation be permitted to discipline its members.42

Disciplinary actions must be based on hearings andthere must be an internal appeal processes withinthe Securities Association. The SEC has oversightof the Securities Association rules, and a membercan appeal to the SEC from a disciplinary decision.Appeal can be had to the federal courts from an SECdecision on any Association’s rules changes or froma disciplinary decision. The National Association ofSecurities Dealers (NASD) is largest self-regulatoryorganization in the United States, with a member-ship that includes virtually every broker/dealer inthe nation that does securities business with the pub-lic. The NASD conduct rules are a massive set.Some are broad, such as, “A member, in the con-duct of his business, shall observe high standards ofcommercial honor and just and equitable principlesof trade.”43 Other rules prohibit very specific con-duct, such as, “No member shall deal with any non-member broker or dealer except at the same prices,for the same commissions or fees, and on the sameterms and conditions as are by such member ac-corded to the general public.”44 The NASD typi-cally disciplines tens of brokers or dealers permonth, with suspensions, disbarments, and fines—far more than the SEC or federal prosecutors.45 Thissystem has been in effect for almost 70 years. Thesecurities self-regulatory scheme is not without its

critics and scandals.46 On the other hand, regulationby those with knowledge of and involvement in theday-to-day real workings of an industry can viewedas effective and popular.

Which model better suits the nanotechnology areais one that policy makers and the public must eval-uate.

THE MYTH OF THE “SHIELD”

Proponents of nanotechnology argue that the bestdefense against an accidental or intentional releaseof nanotechnology would be to build a nanotech-nology shield. That is, one could build defensivenano-machines that would hunt out and destroy anymiscreant nano-machines. The myth of a shield hasbeen shown in at least two different contexts. Un-less nanotechnologists can convince us that theseflaws will not occur in a nanotechnology shield, theidea of a shield may distract policy makers and sci-entists from other avenues that would lead to betterpublic safety.

First, as we know from prior attempts to intro-duce natural predators to combat invading insectsand plants that otherwise have no natural defensesagainst the pest, even where such predators do erad-icate or control these pests, they can become pestsin and of themselves. Often, this problem arises be-cause an environment that has no natural defensesagainst a particular pest probably also has no nat-ural defense against that pest’s predators. In NewZealand, stoats and weasels were imported in an at-tempt to control a rabbit population that was threat-ening to render the Kiwi bird extinct by eating thesame food. Although they ate some rabbits, thestouts and weasels also attacked the Kiwi popula-tion they were meant to protect.47 Similarly, inHawaii, mongooses were imported to try to control

386 Biotechnology Law Report � Volume 22, Number 4

40 15 USC §780(b)(8).41 15 USC §780-3(a).42 15 USC §780-3(b).43 NASD Manual Section 2110.44 NASD Manual Section 2420.45 See generally the notices of such actions by month athttp://www.nasdr.com/2700.asp (last visited June 2, 2003).46 See, e.g., “Securities Markets Regulation: Time to Move toa Market-Based Approach,” Dale Oesterle, Cato Institute, June21, 2000.47 See, e.g., The Land of New Zealand: A Report. SvenMacAller, http://www.stormbefore.com/squatley/newzeal.html(last visited on June 2, 2003).

the rat population. However, because the mon-gooses are active during the day and the rats at nightand they both live in the area and both eat nativebirds or their eggs, the predator is as much a pestas the pest it was supposed to eradicate.48

Second, as Computer Professionals for Social Re-sponsibility (CPSR)49 pointed out during the “StarWars” debates of the 1980s, if a complex and costlyshield stands a fair chance of failure, it may be worsefor the security of a nation than building no shield atall. This outcome has several causes. First, the en-emy is always making benefit and cost calculations.No weapon system is ideal, so a shield is just anothertype of potential “cost” in the calculation for thatweapons system. If the enemy believes that the shieldwill fail even a small percentage of the time, the en-emy may have an incentive to build so many offen-sive weapons that either at least some of them willpenetrate the shield and do significant damage or thesheer numbers of weapons will exhaust the shield. Inthis instance, a shield may actually cause overprolif-eration of arms. Second, a false sense of security be-cause of a costly shield technology may cause mili-tary and political planners (who face a limited budget)to abandon or underutilize lower technology defenseseven though collectively they may be far more ef-fective and robust. Third, confidence in the ability ofa defensive shield may actually cause the nation tobecome more aggressive, as it believes that it can in-flict more damage on its enemy than will be done toit. That is, the United States was more likely to in-vade Serbia, say, to protect human and other inter-ests there (because American defense systems werecorrectly thought to be good enough to prevent dam-age or casualties to Americans by Serbian weapons)than America would be to attack a more technologi-cally developed and savvy nation. Finally, a massiveshield suffers from time lag. Weapons and defensesare an ever-changing game of cat-and-mouse. Unlessone can quickly, inexpensively, and correctly mutatethe nano-machine shield to face new and quicklychanging offensive nano-machines, the shield will beeffective only against what are by then obsolete andunused nano-weapons.

OTHER ISSUES FACINGNANOTECHNOLOGY

Beyond the largest vexing issues facing nan-otechnology, there are other issues policy makers

must face. Some of these are overarching concernsthat affect every stage of nanotechnology develop-ment and deployment. Others primarily affect onlythe funding, development, or deployment stage. Weturn first to the overarching issues.

Issues that affect every stage of nanotechnologydevelopment and deployment

Self-replication: Human bacteria-likenanotechnologies

Because nanotechnology has the power to selfreplicate, it poses dangers that are similar to thoseposed by biotechnology and bioengineering re-search and development but with its own uniquetwists.

In his book Engines of Creation,50 Dr. K. EricDrexler discusses the concept that nano-machinescould be produced to build other nanotechnology.He calls these building machines “assemblers.” Nor-mally, assemblers are nano-machines that build use-ful chemical chains, assemble minute componentsinto a working computer, or the like. However, as-semblers could be programmed to build machinesidentical to themselves, creating more assemblers.That is, nano-machines could become self-replicat-ing.

Some have argued that the danger posed by self-replicating assemblers would be relatively small, astheir only function is to produce more producers.This is not necessarily true. First, as we know fromhuman cancer cells, if producers are inside an im-portant structure (such as a human body, a bridge,or a computer that controls a hospital’s functions)and they reproduce quickly and efficiently, they maycause the surrounding “organ” or structure to ceaseto function properly. Such failure can lead to deathor other harmful consequences.

Further, nano-machines could be programmed todo two functions: replicate a certain number of timesand then move on to do “useful” work. If there wereprogramming or design errors in the instructions forthe “useful” work function, then the danger posedby self-replicating nano-machines could be severe.In the computer field, the analogous problem is the

Biotechnology Law Report � Volume 22, Number 4 387

48 See, e.g., “Polynesian rats,” Mark E. Tobin, in Prevention andControl of Wildlife Damage, USDA, 1994.49 See, e.g., http://www.cpsr.org/publications/newsletters/issues/2001/Spring/index.html (last visited June 2, 2003).50 Engines of Creation, K. Eric Drexler, Doubleday, 1986.

proliferation of computer worms. A worm is a self-contained program that can travel from computersystem to computer system, replicate itself there, dodamage, and then send itself to another intercon-nected computer system. Perhaps the most famousInternet worm was created by Robert Tappan Mor-ris, Jr., on November 2, 1988. It has been reportedthat Morris did not expect the worm to replicate asfast as it did nor to take up so much processingpower on the computers it infested. The problemwas that he set some parameters within the programto the wrong values, and his worm, which was neverintended to do harm, wreaked havoc—all becausehis worm had bugs.51 The problem with complextechnologies is that bugs within them can causethem to behave in an unexpected and destructivemanner. We need to keep this lesson in mind whenwe evaluate the risks of nanotechnology.52

Cloning and nanotechnology

Nanotechnology can be used to clone machinesas well as living creatures. Issues similar to thosecurrently plaguing policy makers about biologicalcloning need to be raised early in the life of nan-otechnology.

Proponents of nanotechnology postulate a worldwhere DNA strands can be custom built by repair-ing or replacing sequences in existing strands ofDNA or even by building the entire strand, fromscratch, one sequence at a time. With enough nano-robots working quickly enough, one could build aDNA strand that will produce a perfect clone. Be-fore Congress now are several bills that would limitor ban cloning of human beings.53 The same issueswill arise, or re-arise, if nanotechnology is success-ful in promoting cloning of DNA segments, cells,organs, or entire organisms.

A prepublication report of the National Academyof Sciences54 highlights the dangers and promise ofcloning. The issues highlighted in the Report echothose that will affect the nanotechnology debate inthis area:

During the committee’s deliberations, fiveoverarching concerns emerged. The first waswhether anything could theoretically go wrongwith any of the technologies. For example, isit theoretically possible that a DNA sequencefrom a vector used for gene transfer could es-cape and unintentionally become integrated

into the DNA of another organism and therebycreate a hazard? The second was whether thefood and other products of animal biotechnol-ogy, whether genetically engineered, or fromclones, are substantially different from thosederived by more traditional, extant technolo-gies. A third major concern was whether thetechnologies result in novel environmentalhazards. The fourth concern was whether thetechnologies raise animal health and welfareissues. Finally, there was concern as towhether ethical and policy aspects of thisemerging technology have been adequately ad-dressed. Are the statutory tools of the variousgovernment departments and agencies in-volved sufficiently defined? Are the techno-logic expertise and capacity within agenciessufficient to cope with the new technologiesshould they be deemed to pose a hazard?55

Even if we assume that the current battles overgenetic bioengineering as a cloning technique will

388 Biotechnology Law Report � Volume 22, Number 4

51 See, e.g., Eisenberg T, Gries D, Hartmanis J, Holdomb D,Lynn MS, Santoro T. The Cornell Commission: on Morris andthe Worm. Commun ACM 1989;32:706–710. Several other ar-ticles in the same issue explore other aspects of the Morrisworm.52 While more far fetched, it is not clear that assemblers couldnot interact with microscopic living organisms. One could imag-ine that a bacterium or other creature could find a way to incor-porate or become symbiotic with the misdesigned assemblers,producing a totally unexpected and bad result. Examples of cap-ture and symbiosis in nature are common. For example, it isthought that mitochondria, which are the energy powerhouses ofcells, were originally bacteria that became permanently capturedby eukaryotic cells millions of years ago. See, e.g., “All FamilyTrees Lead to ‘Eve,’ An African; Scientists Conclude GeneticAnalysis Indicates Common Ancestor 200,000 Years Ago,”Boyce Rensberger, Washington Post, January 13, 1987, A3. Asthe article points out, this is lucky for some scientists, who haveused that fact to show that we all may have a common relative,called “Eve.” Lichens, sharks, and cleaner fish; tick birds on rhi-nos; ox and pecker birds; and termites and their intestinal cellu-lose-digesting flagellates are just a few of the overwhelming ex-amples of symbiosis in nature. The possibility that nano-machinescould become symbiotic with creatures, while remote, cannot becompletely discounted.53 See, e.g., Human Cloning Prohibition Act of 2001, H.R. 2505(introduced 7/16/2001); Human Cloning Ban and Stem Cell Re-search Protection Act of 2002, S. 1893 (introduced 1/24/2002).54 Animal Biotechnology: Science Based Concerns, Committeeon Defining Science-Based Concerns Associated with Productsof Animal Biotechnology, Committee on Agricultural Biotech-nology, Health, and the Environment, Board on Life Sciences,National Research Council, August, 2002.55 Ibid., Executive Summary, page 4.

have been resolved (one way or the other) by thetime nanotechnology perfects its own methods ofcloning, social issues will arise. It is likely that nan-otechnology’s efforts will lead to twists in the as-sumptions that lead to the resolution of cloning is-sues in terms of genetic bioengineering. Policymakers should anticipate, now, that in setting theboundaries for bioengineered cloning, the need toforesee issues that will arise from cloning by nan-otechnology and be ready to reevaluate cloning reg-ulation before nanotechnology perfects its ownmethods of cloning. If we do not anticipate the nan-otechnology problems, the debate will emerge in anenvironment like the current one: one filled with afrenzy and uproar, rather than in an atmosphere ofreflection and deliberateness.

Social policy and law always lag behindscience

It has often been said that law breathlessly triesto keep up with scientific advances. This is likelyto be the case in nanotechnology.

In Chapter 13 of his book, Drexler makes a strongpitch for keeping policy makers out of the debateabout nanotechnology and urges the institution oftechnical panels. He summarizes his argument thisway:

Unfortunately, leaving judgment to expertscauses problems. In Advice and Dissent, Pri-mack and von Hippel point out that “to the ex-tent that the Administration can succeed inkeeping unfavorable information quiet and thepublic confused, the public welfare can be sac-rificed with impunity to bureaucratic conve-nience and private gain.” Regulators suffermore criticism when a new drug causes a sin-gle death than they do when the absence of anew drug causes a thousand deaths. They mis-regulate accordingly. Military bureaucrats havea vested interest in spending money, hiding mis-takes, and continuing their projects. They mis-manage accordingly. This sort of problem is sobasic and natural that more examples are hardlyneeded. Everywhere, secrecy and fog make bu-reaucrats more comfortable; everywhere, per-sonal convenience warps factual statements onmatters of public concern. As technologiesgrow more complex and important, this patterngrows more dangerous.

Some authors consider rule by secretivetechnocrats to be virtually inevitable. In Cre-ating Alternative Futures, Hazel Henderson argues that complex technologies “become in-herently totalitarian” (her italics) because nei-ther voters nor legislators can understand them.

Dr. Drexler sees two flaws with the present pub-lic policy framework. First, regulators have vestedinterests in maintaining their present power and thestatus quo. Second, secrecy and the incentive tocover-up mistakes by “technocrats” harm the for-mation of proper public policy. Thus, Dr. Drexlerproposes “fact forums” of scientific experts to re-place the present public policy framework. He sum-marizes his approach as follows:

We need better procedures for debatingtechnical facts—procedures that are open,credible, and focused on finding the facts weneed to formulate sound policies. We can be-gin by copying aspects of other due-processprocedures; we then can modify and refinethem in light of experience. Using moderncommunications and transportation, we candevelop a focused, streamlined, journal-likeprocess to speed public debate on crucial facts;this seems half the job. The other half requiresdistilling the results of the debate into a bal-anced picture of our state of knowledge (andby the same token, of our state of ignorance).Here, procedures somewhat like those ofcourts seem useful.

Since the procedure (a fact forum) is in-tended to summarize facts, each side will be-gin by stating what it sees as the key facts and listing them in order of importance. Dis-cussion will begin with the statements thathead each side’s list. Through rounds of ar-gument, cross examination, and negotiationthe referee will seek agreed-upon statements.Where disagreements remain, a technicalpanel will then write opinions, outlining whatseems to be known and what still seems un-certain. The output of the fact forum will in-clude background arguments, statements ofagreement, and the panel’s opinions. It mightresemble a set of journal articles capped by aconcise review article—one limited to factualstatements, free of recommendations for policy.

Biotechnology Law Report � Volume 22, Number 4 389

Unfortunately, despite the initial appeal of a sci-entist-driven public policy debate, society haslearned that scientists are not always the best pol-icy makers. A recent article in the Washington Post(May 27, 2002), about flaws in the swine flu vac-cine program of 1976 illustrates this problem. Theswine influenza epidemic of 1918–1919 claimed thelives of between 20 and 100 million people, so whenthe virus reappeared in 1976, public health officialstook quick action. The consensus of the majority ofmedical experts was that an epidemic was likely andthe side effects of a vaccine small. The Post notesthat, “According to various accounts, the idea thata swine flu epidemic was quite unlikely never re-ceived a full airing or a fair hearing, although nu-merous experts apparently held that view. . . . A fewexperts suggested the vaccine be made and stock-piled but used only if there was more evidence ofan epidemic. This was considered but rejected earlyon. The argument was that the influenza vaccine hadfew, if any, serious side effects, and that it wouldbe far easier (and more defensible) to get it into peo-ple’s bodies before people started dying.” That is,the Centers for Disease Control, on the basis of theinput and consensus from medical experts, con-cluded that there was a “strong possibility” of aswine flu epidemic and that “the chances seem tobe 1 in 2.” In fact, the epidemic never emerged, andthe experts were very wrong—the vaccine had se-vere side effects, the worst being a nerve diseaseknown as Guillain-Barré syndrome. The articlenotes, “On Dec. 16, the swine flu vaccine campaignwas halted. About 45 million people had been im-munized. The federal government eventually paidout $90 million in damages to people who devel-oped Guillain-Barré. The total bill for the programwas more than $400 million.” The article ends withlessons from Harvey Fineberg, a former dean ofHarvard’s School of Public Health, “Among them:Don’t over-promise; think carefully about whatneeds to be decided when; don’t expect the con-sensus of experts to hold in the face of changingevents. The biggest, he said recently, was perhapsthe most obvious: Expect the unexpected at alltimes.”

The point here is that although scientific input andexpert panels, perhaps even the “fact forums” pro-posed by Dr. Drexler, are vital to an informed pub-lic policy debate, the politics within academia, thepush for consensus in panels despite minority views,the rapidly changing opinions about scientific issues

on the basis of new evidence, and the fact that oneneeds to expect the unexpected at all times lead tothe need to involve others in public policy debates.Moreover, public policy involves more than scien-tific truth: it involves a balancing of competing so-cietal needs and goals. Broader goals, such as theallocation of scarce resources among competingtechnologies and non-technology needs, the weigh-ing of costs and benefits in pursuing particular proj-ects, whether certain technologies should be regu-lated or banned, whether certain bad activitiesshould be made criminal or should be regulated, andwho should bear the legal liability for damagescaused by the failure of technology, are all issuesthat are beyond the expertise of technical panels butare vital to the conclusion of a rational public pol-icy debate.

On the other hand, it is important for the nan-otechnology community to educate the public andpolicy makers early about important aspects andcharacteristics of nanotechnology so that the debateon public policy is not tainted by those who slantthe scientific facts in the heat of the debate in orderto persuade. Similarly, schools, universities, andgovernments must undertake programs early to ed-ucate themselves and their students or employees onthe science of nanotechnology.

Long-term social effects of the success ofnanotechnology

If its proponents are correct, nanotechnologycould have vast and sudden impacts on our society.Policy makers and society need to consider re-sponses to such profound effects. This paper illus-trates only two examples.

Nanotechnology might increase dramatically thelife expectancy of human beings through diagnos-tic or treatment nano-machines, improved drugs, orDNA repair. This is often seen as a purely positiveoutcome. However, a sudden increase in the life ex-pectancy of a large number of people will likelymean that the carrying capacity of cities, countries,and perhaps even the entire world will be exhaustedin supporting currently living persons. This wouldmean that new births would have to be controlled.Further, longer productive lifespans mean that keypower positions in government, academia, and cor-porations will not be turning over in their normalmanner. As a result, we need to consider the effectson society of a slower turnover of power to the next

390 Biotechnology Law Report � Volume 22, Number 4

generation. One of the great advantages of new chil-dren is that they introduce new ideas and challengeexisting norms. It is said that some of the greatestscientists completed their greatest contributions be-fore the age of 30. Moreover, children grow up toaccept as natural things that their parents found im-possible to live with. For example, racial integra-tion in jobs and the military, and even interracialmarriage, seen a generation ago as an idea that mighttear apart the United States, is now accepted as factby most children.56 Similarly, the use and accep-tance of new technologies, such as computers, is farmore prevalent in children than in their more seniorcounterparts.

If the proponents of nanotechnology are correct,nanotechnology will mean that computers will fi-nally think like human beings. As they envision it,nano-machines will either be small enough to be-come fast enough to break the barrier into “con-sciousness,” or nano-machines will build biologicalcomputers that will mimic the way in which brainsthink and grow. In either case, if they are correct,we need to come to grips with the effects of con-scious computers on society. Will humans find pro-ductive things to do with their time and energies ifcomputers can take over their jobs? Who will con-trol whom? Will computers have the ability to rebelagainst humans? Will computers dominate andeliminate humans and other “living” things? Thesescience-fiction questions will have a greater impactif the most optimistic projections of nanotechnol-ogy come true.57

Issues that affect the development stage

Military funding and directed research candistort scientific research

The military is an enormous funder of scientificresearch. However, the mission of this funding isnot the basic advance of science but the develop-ment of science that can produce weapons, detectthe enemy, or protect troops against an enemy at-tack.

According to the National Science Foundation, inFiscal Year 1990, the defense share of the federalR&D budget authority was 62.6% of the total gov-ernmental R&D budget. In year 2001, it was ex-pected to decrease to 50.1%. Even with this dra-matic decrease, funding by the government ofscientific research is largely devoted to developingmilitary applications.

This is not a problem only in the United States.In an article entitled, “How Should UK Science BeFunded?” it was noted that, “According to latest fig-ures from the OST, one-third of UK governmentfunding for science, engineering and technologycomes from the Ministry of Defence. This amountsto 2.1 billion a year. In comparison, the Dept ofEnvironment, Transport and the Regions is respon-sible for less than 3% of government funding of sci-ence and technology.” Similarly, in India, “Ac-cording to reports issued by the Indian governmentand analyzed in SIPRI Yearbook 1998, the main re-cipient of the government’s scientific largesse hasbeen the Defence Research and Development Or-ganisation (DRDO), which performs roughly 85percent of India’s military research and develop-ment. The organization received 15 billion rupeesin fiscal year 1996–97, the last for which compre-hensive statistics are available.”

Military funding poses both personal ethical andsocietal challenges. Personally, scientists involvedin nanotechnology need to be aware of, and cometo grips with, the fact that their own research maylead to the production of weapons of mass destruc-tion. A number of scientists involved in the devel-opment of the nuclear bomb, in retrospect, foundthat knowledge hard to live with.58

Military funding can also have distorting effectson the progress of science as a whole. Scientistsneed to gain funding for their research. They needto prepare grant proposals that win funding ap-proval. Thus, they naturally tailor their proposalsand research to areas that will catch the attention ofthe granting organization. In the case of the mili-tary, they need to slant proposals to weapons de-velopment. In some cases, the funding organizationtells the researchers what types of proposals they arelooking for. In such cases, it is obvious how scien-tists must alter, or at least tailor, the focus of theirresearch to meet the goals of the request for pro-posals. In other cases, the proposals are more open,

Biotechnology Law Report � Volume 22, Number 4 391

56 “Biracial Couples Report Tolerance; Survey Finds Most areAccepted by Families,” Darryl Fears and Claudia Deane, Wash-ington Post, July 5, 2001, A1; “Racial Divide in Sports Doesn’tMatter to Athletes; They Say that Playing Brings People To-gether,” Camille Powell, Washington Post, June 21, 2001, T10.57 See, e.g., “Why the future doesn’t need us,” Bill Joy, Wired,8.04, April 2000.58 See, e.g., misgivings of Werner Heisenberg and speeches ofJ. Robert Oppenheimer.

but again, the scientist must write to his or her au-dience and propose projects that will win approval.In many cases, one can look at proposals that havewon in the past and follow that well-trodden path.In this way, even when there is an open call for pro-posals, the proposals that are submitted are distortedby the knowledge that they being submitted to a mil-itary organization.

But this structural distortion is not unique to mil-itary funding: it happens in the growing area of di-rected research. There has been an increase in cor-porate and other directed research. This is notnecessarily a bad thing: particularly with govern-mental sources of R&D and basic scientific researchbeing reduced, corporate funding of basic science iswelcome. Moreover, a partnership between thosewith important scientific expertise and those whoare producing actual products and services can yieldsignificant results.59 On the other hand, certain di-rected research by tobacco companies has been citedas an example of corporate research money beingused to try to advance bad scientific positions in or-der to ward off or counter commonly held scientificprinciples.60 The tendency for directed money todistort otherwise-objective views cannot be denied,just as it does when it comes from the military orother sources (including nonprofit advocacy groups)that seek a particular outcome for the research.

Society must come to grips with the good and badeffects of directed research. Although this is not atopic unique to nanotechnology, it is one that canhave the effect of distorting or inappropriately redi-recting science onto paths that are not in society’sbest interests.

Inherent conflicts of interest betweenresearch or commercial exploitation anddisclosure or sharing of results

Recently, there have been a number of scandalsinvolving failure to timely report incidents in hu-man clinical research. For example, WashingtonPost reported that “The University of Pennsylvaniaannounced yesterday that its gene therapy institute,which has been an international leader in the cut-ting-edge field of medical research, will no longerexperiment on people.”61 The Post noted, “The uni-versity’s action came after the Food and Drug Ad-ministration found that Wilson had not properly re-ported the deaths of experimental animals or seriousside effects suffered by volunteers who preceded

Jesse Gelsinger, the Tucson teenager who died Sept.17 after undergoing an experimental therapy for arare metabolic disorder.” These incidents are notlimited to the University of Pennsylvania. The samearticle continues, “In addition to Penn’s problems,the field—which tries to cure disease by giving peo-ple healthy copies of “disease” genes—has beenrocked by revelations that researchers elsewhereweren’t properly reporting the deaths and illnessesof hundreds of volunteers to the National Institutesof Health as required by federal regulations. . . .Most recently, the FDA shut down four gene ex-periments by a prominent researcher at Tufts Uni-versity and cited him for numerous safety lapses, in-cluding the failure to tell his own institution aboutthe death of a volunteer and the inclusion of patientswho did not qualify and may have been harmed bythe experimental treatment.” Six months after thisarticle was published, the Washington Post reportedthat “A Harvard-affiliated hospital in Boston qui-etly suspended a gene therapy experiment last sum-mer after three of the first six patients died and aseventh fell seriously ill, previously unreleased re-search records show. Richard Junghans, the HarvardMedical School researcher who led the study,blames the problems on a series of tragic coinci-dences that were mostly not related to the treatment.But the federal committee that oversees gene ther-apy had no chance to question that conclusion—orshare it with other scientists working on similar ex-periments—because Junghans did not report thedeaths or illness to the National Institutes of Healthwhen they occurred, as required by federal regula-tions.”62

These incidents illustrate the point that re-searchers, commercial and academic alike, have in-

392 Biotechnology Law Report � Volume 22, Number 4

59 See, e.g., “At Kansas State, Seeking Patents, with Hopes ofProfits Pending, Turns Donated Rights into Products, Compa-nies and Jobs,” Robert E. Pierr, Washington Post, June 8, 2002,A3.60 See, e.g., “The Smoke You Don’t See: Uncovering TobaccoIndustry Scientific Strategies Aimed against Environmental To-bacco Smoke Policies,” Am J Public Health 2000;91:1419–1423, September 2000; Dearlove JV, Bialous SA, GlantzSA. Tobacco industry manipulation of the hospitality industryto maintain smoking in public places. Tobacco Control 2002;11:94–104.61 “Penn Ends Gene Trials on Humans,” Deborah Nelson, RickWeiss, Washington Post May 25, 2000, A1.62 “Earlier Gene Test Deaths Not Reported; NIH was Unawareof ‘Adverse Events’,” Deborah Nelson, Rick Weiss, Washing-ton Post January 31, 2000, A1.

herent conflicts of interest that cause them to fail toconduct research properly, to report failures, and toadmit mistakes. The proponents of nanotechnologyhave argued that scientists can police themselvesand can be trusted to adopt and use safe experi-mental methods and to report incidents. Society haslearned from experience that even the best-inten-tioned researchers do not always follow safe proto-cols and report adverse events.63

Related to this conflict is the conflict that arisesduring commercial exploitation of a new technol-ogy. Commercial exploitation of science inherentlyrequires that the researcher keep confidential theoutcome of his or her research in order to providethe researcher (or his or her employer) a competi-tive advantage over competitors and to keep com-petitors from “free-riding” on the results of this veryexpensive research. This is not inherently a badthing. Being able to keep important developmentssecret until they are ready to be marketed and soldto the general public gives researchers an importantincentive to continue to do leading-edge research.On the other hand, as seen in the medical field, thishas sometimes meant that drug side effects or badinteractions are not timely disclosed to regulatoryagencies or the public. The same conflict could af-fect nanotechnology research.

Finally, there is a related conflict of interestproblem where the scientist has a financial stakein the outcome of the research. For example, as theWashington Post reported recently on its front pageon June 30, 2002, “One of the nation’s largest can-cer centers enrolled 195 people in tests of an ex-perimental drug without informing them that theinstitution’s president held a financial interest inthe product that stood to earn him millions. Thetests at M.D. Anderson Cancer Center in Houstoninvolved Erbitux, the controversial cancer drugthat is at the center of broad investigations in NewYork and Washington. Most of the patients, whowere quite ill by the time they enrolled in the tests,have died. The cancer center, a unit of The Uni-versity of Texas system, has since acknowledgedthat it should have informed the patients of the con-flict of interest involving its president, JohnMendelsohn. It has recently adopted policies to en-sure that patients are told ahead of time if Mendel-sohn or the cancer center itself has a financialstake. Ethicists say that such conflicts of interestpose risks to patients and to the integrity of scien-tific studies.”64

Conflicts of interest are not new, but they do posea societal risk. Policy makers and regulators need tobe proactive in evaluating ways to ensure that theseconflicts of interests do not keep societal risks hid-den from them.

Issues that affect the deployment stage

Workplace issues

As with other technologies, workers in assemblyplants will be exposed to byproducts of the manu-facturing process. These byproducts could includetoxic chemicals that are used to produce the nan-otechnology, as well as unusable microscopic piecessuch as nano-wires that escape the manufacturingenvironment and float free in the air.

As reported in Micro magazine, “Although SIAhas touted its annual U.S. government ranking inthe top 5% of all U.S. industries in worker safety,critics argue that exposure in the cleanroom tochemicals such as arsine, benzene, and HCl heightenthe risk of cancer and miscarriages.” The articlegoes on to say:

In general, electronic computer equipmentis a complicated assembly of more than 1,000materials, many of which are highly toxic,such as chlorinated and brominated sub-stances, toxic gases, toxic metals, photo-activeand biologically active materials, acids, plas-tics and plastic additives. The list of toxic ma-terials in computer components also includeslead and cadmium in computer circuit boards,lead oxide and barium in computer monitors’cathode ray tubes, mercury in switches and flatscreens, and brominated flame retardants onprinted circuit boards, cables and plastic cas-ing. Comprehensive health impacts of the mix-tures and material combinations in the prod-ucts are often not known.65

Biotechnology Law Report � Volume 22, Number 4 393

63 See also “Science Breaks Down When Cheaters Think TheyWon’t be Caught,” Sharon Begley, Wall Street Journal, Sep-tember 27, 2002, B1, on why, despite the fact that scientificfraud seems to be a counterproductive and irrational activity, itseems to occur with some regularity.64 “A Hospital’s Conflict of Interest: Patients Weren’t Told ofStake in Cancer Drug,” Justin Gillis, Washington Post, June 30,2002, A1.65 “Study Results Prompt SIA to Examine Whether Fab Chem-icals Imperil Workers’ Health,” 2001 Micro, April 2002.

There is an additional risk in nanotechnology: mi-croscopic pieces of assemblies can break off andfloat in the air. The later risk of hazardous airbornemicrofibers is illustrated by the societal problemsthat arose from asbestos. “An estimated 1.3 millionemployees in construction and general industry facesignificant asbestos exposure on the job. Heaviestexposures occur in the construction industry, par-ticularly during the removal of asbestos during ren-ovation or demolition. Employees are also likely tobe exposed during the manufacture of asbestos prod-ucts (such as textiles, friction products, insulation,and other building materials) and during automotivebrake and clutch repair work.”66 Ashahi Weekly, inan article entitled “Asbestos Deaths Seen Likely toSoar,” made it clear this is a problem in Japan (andother nations), not just the U.S. In Japan, the articlestated, “An estimated 100,000 people will die fromcancerous lung tumors over the next four decadesdue to past exposure to asbestos, researchers say.”

While nanotechnology is seen by its proponentsas the ultimate “clean” technology, it will presentrisks to workers in the near term and may alwayspresent workplace safety issues. Society needs tounderstand that these risks exist and deal with them.

Nanotechnology as a police/big brother tool

Nanotechnology can penetrate places and deviceswithout detection. It can detect and recover infor-mation that people otherwise have the ability to keepsecret. As such, nanotechnology has the potential tobe a massive engine of police repression or over-sight.

The miniaturization of electronics has opened upnew dangers to privacy and human rights. Smallcameras can be implanted in places once incon-ceivable. Electronic bugs can listen in quiet or verynoisy places and transmit the data with very lowelectronic emissions. Today, the normal means ofimplanting such devices requires a human being toimplant the device in the hostile location. In the fu-ture, with nano-machines being able to move ontheir own, the implantation of ever-smaller devicesmay become possible.

In another vein, proponents of nanotechnologynote that with smaller integrated chips, the powerof computers will greatly increase. As a result, en-crypted private messages will be more vulnerable tobeing decrypted by unauthorized persons. Obvi-ously, where such a person is a law enforcement of-

ficial working under a court search warrant, that isa positive development. Where it is a competitor ofa business or a hacker it is a social concern.

Software issues and nanotechnology

Some of the plans for nanotechnology involve co-ordination of the activities of huge numbers of nano-machines. This coordination will be done by com-puter programs. These complex programs mayreside inside a single computer that directs the ac-tions of the dispersed nano-machines or may involvethe coordination of the actions of millions of inde-pendent micro-modules, each with its own copy ofa portion of the software. In either case, softwareplays a crucial role in the operation and coordina-tion of these nano-components. We have learnedthat software is inherently buggy and susceptible tomassive catastrophic failure.

As David Parnas, a member of CPSR, and otherspointed out during the Star Wars debates of the1980s, computer scientists have long known thatsystems of great length and complexity (the U.S.Navy’s AEGIS combat software contains 2 millionlines of code; the Star Wars program was estimatedto need between 7 million and 60 million lines ofcodes) are likely to be filled with bugs that createunexpected and often catastrophic failures.67 Test-ing and debugging software is more an art than ascience, and even well-meaning expert computerprofessionals are unable to predict reliably what ac-tions and results may occur from bugs in code.68 Insome cases, bugs lead to shutdown of systems,which in the case of many nanotechnology applica-

394 Biotechnology Law Report � Volume 22, Number 4

66 “Asbestos.” OSHA Website, http://www.osha-slc.gov/SLTC/asbestos/ (last visited June 2, 2003).67 These size estimates, as well as both sides of the argumentsrelated to software reliability, were summarized and analyzedin SDI: Technology Survivability and Software, Office of Tech-nology Assessment, Congress of the United States, May 1988.That report concluded “The nature of software and experiencewith large, complex software systems indicate that there willalways be irresolvable questions about how dependable [ ] soft-ware would be and about the confidence the United States couldplace in dependability estimates” (See the Report at page 4).For a recent summary of these arguments, see “National Mis-sile Defense: The Trustworthy Software Argument,” WilliamYurcik, CPSR Newsletter, Volume 19, Number 2, Spring 2001,and other articles in that issue.68 An extensive discussion of software failures and the reasonsfor them, as well as overlying ethical considerations, can befound in Chapter 5 of Computer Ethics (second edition), TomForester and Perry Morrison, 1994.

tions may be an acceptable outcome. On the otherhand, sometimes, the unexpected shutdown of sys-tems (such as in the case of a nanotechnology shield,a nanotechnology detection system, or a manufac-turing process) can lead to unacceptable outcomes.Moreover, not all bugs lead to shutdown. Many per-mit the system to continue to operate, but the ac-tions the system takes and the results or outputs areflawed, even dangerous.

Where nanotechnology is deployed to interactwith human bodies, plants, and animals in open en-vironments and to control vital physical machines,software bugs present a societal risk that must beweighed in the implementation of any nanotechnol-ogy project.

Nanotechnology incident reporting statutes

For some specific diseases and for some kinds ofincidents, current laws require prompt reporting toa central authority. Today, no such laws exist withrespect to nanotechnology.

There are a host of statutes in various states thatrequire health workers to report incidents of certaindiseases in humans and other animals, such as AIDS,anthrax, botulism, cholera, gonorrhea, rabies (hu-man), syphilis, and tuberculosis. There are a host ofstatutes that similarly require reporting of certain in-cidents, such as child abuse. The OSHA regulationsrequire employers to keep records and report safetyincidents in the workplace. Since 1975, the FDA hasmandated reporting of major blood transfusion reac-tions. Reporting of certain types of incidents at nu-clear plants is mandatory. Incidents that occur duringhuman testing are required to be recorded promptly.The reasons for such reporting requirements are var-ied, but generally, they try to give a warning to offi-cials early enough in the process that they can seepatterns emerging, take actions to prevent the spreadof the problem, quickly track down the origins of theincident, mobilize diverse resources at appropriatelevels, and keep long-term statistics on each type ofdisease or incident. Laws or regulations that requirereporting of certain types of nanotechnology inci-dents may need to be considered.

PATENT AND INTELLECTUAL PROPERTYPROTECTION OF NANOTECHNOLOGY

Nanotechnology may seem to fall logically withinthe existing protections of patent, copyright, and

trade secret law. However, it is important to con-sider the benefits and costs of a sui generis form ofprotection for this new intellectual property.

Traditionally, patents have been the major type ofintellectual property protection for physical inven-tions. Patents protect innovative inventions, pro-cesses, and designs. Part of what makes patent pro-tection so powerful is: (1) patents protect a very widerange of subject matter; and (2) innocent infringe-ment is not a defense. In other words, one can pro-tect, not only mechanical devices, but also formulasfor drugs, genetic sequences, software, business pro-cesses, and even the means to manufacture devices,under patents. Second, unlike copyright, for exam-ple, the independent creation of the same or a simi-lar invention or process is still an infringement of thepatent and subjects the later inventor and all thosewho use the invention process to damage claims.

Patents have been controversial in the software andbusiness process area. The controversy swelled in theUnited States in the late 1990s, after the SupremeCourt ruled, first, that software was patentable subjectmatter69 and later courts ruled that business methodswere patentable.70 In recent years, the controversy hasquieted somewhat within the United States but is be-ginning to engender passions in Europe, after the Eu-ropean Union released a proposal for patenting soft-ware.71 Among the reasons people cite for objectingto patent protection for software are: (1) that patentsprotect inventions from the date of filing, but thepatent office does not make the application for a patentpublicly known for a period that may stretch to up to18 months; (2) that the cost of obtaining a patent andthe cost of defending against even an invalid claim ofinfringement can run in the tens to hundreds of thou-sands of dollars; (3) that software and the Internet donot need patents to spark innovation and thrive; and(4) that patents increase barriers to entry because theycreate monopolies within the scope of the patent.72

Biotechnology Law Report � Volume 22, Number 4 395

69 Diamond v. Diehr, 450 U.S. 175 (1981).70 State Street Bank & Trust Co. v. Signature Fin. Group,Inc., 149 F.3d 1368 (Fed. Cir. 1998), cert. denied, 119 S. Ct.851 (1999).71 Directive on the Patentability of Computer-ImplementedInventions, 20.02.2002 COM(2002) 92 final 2002/0047(COD).72 See, e.g., Should Patents be Granted for Computer Soft-ware or Ways of Doing Business: The Government’s Con-clusions, U.K. Patent Office, March 2001; James Gleick,“Patently absurd,” New York Times Magazine, March 13,2002; Robert Lemos, “Patents, lawsuits plague the Net,” ZD-Net News, March 30, 2000.

Copyright protects a particular tangible expres-sion made (or designed) by a human being fromunauthorized copying, distribution, the making of“derivative works,” and public display and perfor-mance. Independent creation is a defense—that is,to prove infringement, one must show that the in-fringer had access to the work and copied the work.Copying can be shown indirectly if placing the twocreations side by side convinces the judge or jurythat copying was likely. There are a series of “fairuse” defenses to copyright infringement. Also,copyright cannot be claimed in the utilitarian ele-ments of a work if that is the only way to do some-thing. That is, under the “merger doctrine,” whileone can protect the design of a fanciful Coke bot-tle, one cannot copyright the shape of a shoe box,because a shoe box is likely to be in the shape of arectangular cube in order to accommodate shoes thatare much longer than wide, with a lid to open, andin that shape to permit stacking of the boxes. Thus,a chemical compound designed by a human maywell not be able to be copyrighted, as its utilitarianfunction dictates its shape. Similarly, while onemight be able to copyright a nano-sculpture, onemay well not be able to copyright a nano-robot’sshape that is dictated by its function.

Trademark protects any logo, shape, color, slo-gan, or other graphic, words, sounds, etc. that iden-tify the attached product or service as coming froma single source. A trademark indicates that a trustedsource oversees the quality of the product or ser-vice. That is, McDonald’s does not trademark itshamburger per se; it trademarks the name “Mc-Donald’s” as an indicator of the entity that ensures

the quality of the burger. That identifying source canbe protected under trademark by McDonald’s evenif it later ventures into other fields such as hotels,software, or airplanes.

In sum, the difference between patent, copyright,and trademark is that a patent protects a particularinvention (or device), process, or design; copyrightprotects a particular audiovisual expression (such asa novel, software, or design); and trademark pro-tects the manufacturer’s (or other business’s) tradenames.

A series of sui generis or neighboring rights havealso been created in recent years. Among these isthe European Union’s database protection scheme.In brief, factual databases are protected for a shortperiod (15 years) only against the unauthorized andsubstantial extraction of the data. Moral rights are asimilar example. While moral rights differ greatlybetween countries, in essence, they protect an au-thor by requiring that the work produced by that au-thor be attributed to him or her and not be edited oraltered without the author’s permission, even if theauthor is no longer (or never was) the owner of thecopyright in the work.

As nanotechnology progresses, the weaknessesand strengths of the patent/copyright/trademarkregime have to weighed against the creation of ascheme that is tailored to the peculiarities of nan-otechnologies. The same debate that is now playingout in the patenting of business processes, software,or biotechnologies will play out with nanotechnol-ogy. We need to look at the resolutions of those de-bates in evaluating what protection should begranted to nanotechnology.

396 Biotechnology Law Report � Volume 22, Number 4