ńňđ. 9
(âńĺăî 23)



For the moment, the FFDOP remains a voluntary programme, and any
pilot who volunteers and completes the training course is considered quali-
fied (subject to a pre-determined maximum number of pilots who may be
deputized).35 The Act grants air carriers total immunity from liability for
damages in actions filed in federal or state courts arising out of an armed
pilot’s use or failure to use a firearm.36 However, the Act only partially
protects the pilots themselves – an armed pilot can still be liable in cases
involving gross negligence or wilful misconduct.37
One argument in favour of armed pilots is that there is less risk of
passengers being accidentally shot, given that pilots are largely confined to
cockpits. Their only responsibility would be to defend the cockpit against
intrusion, a task facilitated by the requirement for hardened cockpit doors.38
However, many countries, airlines and crew unions have strenuously
opposed proposals to arm flights crews with lethal weapons or to require
them to undergo training in the use of lethal force.39 The fear is that armed
pilots, like armed air marshals, introduce the risk of stray bullets being fired
in the cockpit which can jeopardize electrical or navigation equipment.40
Arming pilots could also have detrimental consequences if a fatigued
pilot’s judgment is impaired by long flying shifts. The lack of sleep can
negatively affect perception and reaction time and contribute to the risks of
pilot error.41 Also, it would be unwise to distract a pilot or to have his
attention diverted from flying – in principle as well as in practice, a pilot
should not have to choose between his duties as a pilot and his duties as a
federal law enforcement officer.42 The defence of the aircraft should properly
be left to air marshals. At the same time, allowing pilots to bring guns into
airports and onto aircraft runs contrary to the overall security objective,
which is to prevent potential terrorists from gaining access to a weapon.43
The economic costs of a pilot arming and training programme can also be
substantial, with the costs likely to be passed on to consumers.44
Both ICAO and IATA have instead called for flight crews to be trained in
non-lethal forms of self-defence and/or that non-lethal protective devices be

APATA, s. 1402(a). The FFDOP has now been extended to cargo pilots and other
Renna, ‘Fire in the Sky’, at 867. 36 APATA, s. 2(h)(1). 37 APATA, s. 2(h)(2).
Lott, ‘Marshals are Good’.
See, e.g., IATA’s response through GASAG and Renna, note 24 above, at 860.
Bartholomew Elias, ‘Arming Pilots against Terrorism: Implementation Issues for the
Federal Flight Deck Officer Program’, Report for Congress, Congressional Research
Service, Library of Congress, Order Code RL31674 (25 March 2003), at 6.
Renna, ‘Fire in the Sky’, at 870. 42 Ibid., at 874. 43 Elias, ‘Arming Pilots’, at 6.
Miller, ‘The ‘‘Cost’’ of Securing’, at 434.

made available in the cabin area (e.g. stun guns, pepper sprays, etc.).45
Overall, it is imperative for ICAO and IATA members to achieve some
form of negotiated compromise on this issue, again flexibly allowing for
differing perceptions of risks in different parts of the world.
The other major security measure enacted after 9/11 is the fortification of
cockpit doors. Section 104 of ATSA authorizes the relevant government
agency to require that cockpit doors have locks and can be fortified so they
cannot be forced open from the passenger compartment.46 It also requires
that the door remain locked during flight and that no member of the flight
crew has a key unless assigned to the flight deck.47 Furthermore, ATSA
provides the authority to take other necessary action, including modification
of safety and security procedures and flight deck redesign.48 ATSA also
provides for funding to airlines to fortify their cockpit doors.49
ICAO, through amendments to the Chicago Convention’s Annex 6, has
also thrown its weight behind the fortification of cockpit doors.50 This has
been one area where it has been thought that some measure of global
harmonization is desirable and viable. ICAO’s cockpit door measure was
largely influenced by the US provisions, with the FAA working closely with
ICAO to develop the standards after they had entered into force domestically
and for all foreign airlines entering the US.51 By 1 November 2003 all passenger-
carrying planes above a certain mass and carrying more than sixty passengers
must be equipped with cockpit doors designed to ‘resist penetration by small
arms fire and grenade shrapnel, and to resist forcible intrusions by unauthor-
ized persons . . . [and] be capable of being locked and unlocked from either
pilot’s station’.52 In addition, the cockpit door is to be locked from embarka-
tion to disembarkation.53
On its part, IATA has made similar recommendations and supports the
employment of advanced cockpit door technology. Meanwhile, the EU has
called on its member states to support the ICAO amendment to Annex 6.54
Overall, the costs of hardening cockpit doors have been considerable. In
September 2003 the TSA announced that it had reimbursed fifty-eight
domestic air carriers a total of $100 million for the cost of reinforcing their
cockpit doors. This was in addition to the $97 million already reimbursed by

See, e.g., ICAO’s Amendment 27 to Annex 6, para. 13.4.1. 46 ATSA, s.104(1)(a)(1)(B).
Ibid., s.104(1)(a)(1)(C), (D). 48 Ibid., s.104(1)(a)(2). 49 Ibid., s.118(c)(1).
Amendment 27 to Annex 6. The requirement applies to aircraft heavier than 45,500 kg or
seating more than 60 passengers.
See William Karas and Carol Gosain, ‘Recent US Regulation of Foreign Airline Practices:
Impermissibly Unilateral or Not?’ (2002) 16 Air & Space L. 4, at 6.
Annex 6, Amendment 27, para. 13.2.2. 53 Ibid., para. 13.2.3(a).
Communication from the Commission, ‘Co-ordinated answer by Member States to the
ICAO document AN 11/1.3.16–02/23 on new protection standards for flight crew com-
partment doors,’ Com (2002) 444(01).

the FAA earlier.55 In the future it is likely that there will also be costs to
maintain, replace and reinstall doors.56

B. Airport security – baggage and passenger screening
On the ground, enhanced baggage and passenger screening are the most
significant new measures employed.57 In the US, ATSA has federalized bag-
gage, passenger, mail and cargo screening tasks.58 All baggage, including
checked-in baggage, are to be screened. ATSA also laid down a requirement
for explosive detection systems (EDS) to be installed in all US airports by 31
December 2002, and for all checked baggage to be screened by these sys-
tems.59 Employees screening baggage must be US citizens and possess a high
school diploma or equivalent, or have experience deemed sufficient to per-
form screening duties.60 Screeners must also have a satisfactory score on a
federal selection examination and possess the basic aptitudes and physical
abilities to use the screening equipment.61
The federalization of screening responsibilities was brought about by the
belief that better-paid and better-trained personnel would produce better
results,62 and that such responsibilities could only be carried out by properly-
selected US citizens. Implicit in this requirement was the belief that non-US
citizens posed a greater threat to domestic security than citizens, particularly
if these were illegal aliens. Predictably, the requirement has resulted in the
displacement of thousands of non-citizen airport screeners in the past few
years.63 The costs of the additional screening measures have also been extre-
mely high.64 At the same time, the EDS machines have been shown to be

TSA, Press Release, ‘TSA Approves $100 Million to Reimburse Airlines for Reinforcing
Cockpit Doors’ (24 September 2003), http://www.tsa.gov/public/display?content ¼
0900051980056949. See also note 101 on costs.
The initiative has been criticized as ineffective. In one incident, an overnight cleaning crew
at Dulles Airport near Washington, D.C. apparently rammed a drinks cart into one of the
new doors on a United Airlines plane, breaking the door off its hinges, see Lott, ‘Marshals
are Good’.
See generally Jack H. Daniel III, ‘Reform in Airport Security: Panic or Precaution?’ (2002)
53 Mercer L. Rev. 1623, and Jamie L. Rhee, ‘Rational and Constitutional Approaches to
Airline Safety in the Face of Terrorist Threats’ (2002) 49 DePaul L. Rev. 847.
ATSA, s.110. 59 Ibid., ss.110(b)(2) and 110(c).
Ibid., s.111(a)(2). See also Francine Kerner and Margot Bester, ‘The Birth of the
Transportation Security Administration: A View from the Chief Counsel’ (2002) Air &
Space L. 20, at 22.
ATSA, ss. 110 and 111. 62 Miller, note 4 above, at n.144.
See Paul S. Dempsey, ‘Aviation Security: The Role of Law in the War Against Terrorism’
(2003) 41 Colum. J. Int’l L. 649, at 716.
For the 2005 financial year, the TSA has requested $4.8 billion to cover screening costs.
The cost of an EDS machine is approximately $1 million, with between $700,000 and

error-prone, with an initial figure of 22 per cent in false-positive records.65
This problem can lead to inconvenient delays as well as potential invasions of
privacy for individuals whose bags are inaccurately singled out by the EDS.
In the EU, Reg. 2320/2002 requires all checked-in baggage to be placed in
areas where only authorized persons have access,66 and that such baggage be
protected from any unauthorized interference from the time of check-in to
departure.67 If a passenger has checked in baggage but does not board the
flight, such baggage will be removed from the aircraft, as had been common
practice in most states even before 9/11.68 Checked-in baggage must be
screened by one of several methods, including hand search, conventional
X-rays, EDS, and Trace Detection Equipment.69
ICAO has also laid down improved screening procedures. Pursuant to
Amendment 10 to Annex 17, states are required to take action to detect
weapons, explosives or other dangerous devices.70 States are to ensure that
by the year 2006 all checked-in baggage are screened at airports, and that
screeners are subjected to stringent background checks and selection proce-
dures.71 While such requirements may easily be met by the developed states,
there will presumably be difficulties for airports in developing countries. The
need for flexibility is recognized by ICAO itself, which is not attempting to lay
down specific standards or methods for baggage screening. Due to the
unreliability of EDS measures, it may well be advisable for states to adopt
the EU’s practice of combining different baggage detection systems, includ-
ing the use of hand searches. IATA’s GASAG recommendation for countries
to work together and share R&D costs to develop better EDS technology
should also be taken up.
What has become more far-reaching and controversial post-9/11 has been
the screening of passengers.72 This now goes beyond conventional metal
detectors and body searches to include the collection of passenger data, the
employment of biometric technology and the use of data collected through
various means to screen passengers through a process of profiling. In

$1 million needed annually for operation and maintenance. See also note 101 for a brief
assessment of the total costs of aviation security.
Dempsey, ‘Aviation Security’, at 723. 66 Annex, Part 5, s.5.1(1)(c). 67 Ibid., s.5.3(1).
Ibid., s.5.1(2).
For the US position, see Frank Costello, ‘The New Federal Approach to Transportation:
The First Three Months of the Transportation Security Act and the Transportation
Security Administration’ (2002) 14 DePaul Bus. L. J. 333, at 336.
See generally Dempsey, ‘Aviation Security’, at 678.
Ibid., at 678–79. On its part, IATA supports the ICAO’s standards through its RSSs,
particularly Recommendations 8 to 10.
For screening generally, see Jonathan Miller, ‘Search and Seizure of Air Passengers and
Pilots: The Fourth Amendment Takes Flight’ (1994) 22 Transp. L.J. 199.

particular, the TSA is now authorized to provide for the use of biometrics to
identify passengers who pose a threat to aircraft security and to subject them
to additional screening.73
Pursuant to the USA Patriot Act,74 the Department of Homeland Security
has incorporated the use of biometric technology into the US Visitor and
Immigrant Status Indicator Technology Program (US-VISIT). In addition,
biometric technology has been considered for use in IATA’s Simplified
Passenger Travel (SPT) programme, which is meant to alleviate passenger
clearance at airports. Under the programme passengers can simply swipe
their cards which hold relevant information in order to attain instant clear-
ance every time they travel.75
Pursuant to the US-VISIT programme, the entry and exit of all non-US
citizens at airports will be tracked through the use of inkless fingerprints
and digital photographs. The information will be used to verify visitors’
identities and to match these against ‘watch lists’ of known high-risk
individuals.76 In addition, the Patriot Act mandates that all countries on
the Visa Waiver Program (which permits entry into the US without visas) are
to issue their citizens with machine-readable travel documents (MRTDs)
carrying biometric identifiers that comply with ICAO standards by 26
October 2005.77

ATSA, note 6 above, s. 109(a)(7).
Uniting and Strengthening America by Providing Appropriate Tools Required to
Intercept and Obstruct Terrorism Act, H.R. 3162, 107th Cong., 2001, s.414(b)(1). See
also the Immigration and Naturalization Service Data Management Improvement Act of
2000 (DMIA), Public Law 106–215 (2000), 114 Stat. 339 (codified as amended at 8 U.S.C.
See generally Ruwantissa Abeyratne, ‘Attacks on America: Privacy Implications of
Heightened Security Measures in the United States, Europe, and Canada’ (2002) 67
J. Air L. & Com. 83, at 83–84. For SPT, see http://www.simplifying-travel.org/.
A measure of tit-for-tat was reported by irate states like Brazil, which reacted to the US
programme by announcing its own fingerprinting of US nationals entering Brazil.
The original deadline was 26 October 2004. However, it has been reported in recent
months that countries on the visa waiver programme (VWP), including allies like the
European states, do not expect to be able to issue passports with biometric identifiers by
26 October 2004. The US House of Representatives’ extension on 16 June 2004 of the
deadline by one year was approved by the Bush Administration in August 2004. However,
doubts remain over whether biometric identifiers can be developed in time by October
2005, see Face Recognition Passports Expected by December, http://www.cnn.com/2004/
TECH/06/15/face.passport/. Meanwhile, visitors from VWP countries will continue to
enjoy visa-free travel but will have to be fingerprinted and photographed from
30 September 2004 pursuant to the US-VISIT programme. All passports from VWP
countries must also be machine-readable from October 2004. Note that Canada is not
part of the visa waiver programme, and Canadians do not generally require visas to enter
the United States.

ICAO-recommended standards for biometrics are laid out in its Technical
Report on MRTDs, which drew heavily from the US experience.78 The
standards take into account such factors as global inter-operability, unifor-
mity and technical reliability. Facial recognition was identified as the pre-
ferred globally inter-operable biometric, though states can elect to
supplement this with fingerprint and/or iris recognition. On its part, the
EU has yet to enact any laws concerning biometrics in MRTDs. However, it
has considered a proposal on common standards for biometrics, taking into
account the US deadline and the ICAO standards.79 Thus, it is likely that
some form of global harmonization in relation to biometric identification for
travellers will be in place soon.
At the same time, passenger information is being actively collected in the
US pursuant to ATSA.80 Thus, all carriers operating passenger flights into
the US must electronically transmit a ‘passenger and crew manifest’ to the
relevant authority before landing in the US. ATSA permits such information
to be shared with other federal agencies for the purpose of national security.
The list of information required includes the date of birth of passengers and
crew and the number and country of issuance of their passports. Carriers are
also required to make passenger name records (PNRs) available to the Bureau
of Customs and Border Protection (constituted under the Department of
Homeland Security) upon request.81
In addition, US legislation has provided for another database system
known as Advanced Passenger Information (API) as a requirement for all
carriers flying into the US.82 The API system works on the basis of each
passenger’s information being transmitted to immigration and customs
authorities in advance of his/her actual arrival. The information would then
be used to determine if the passenger should be admitted to the destination
state.83 Concurrently, TSA is required to keep a ‘watchlist’ of names posing an
immediate threat to passenger safety in the air.
Yet another US initiative is the Computer Assisted Passenger Prescreening
System (CAPPS II), authorized by ATSA.84 This programme seeks to authen-
ticate travellers’ identities and perform risk assessments to detect individuals
who may pose a terrorist-related threat, and involves asking passengers
for certain information well before the flight, typically at the time of the
reservation of air tickets. Individuals will be categorized or profiled according

On 28 May 2003, ICAO adopted a global, harmonized blueprint for the integration of
biometric identification information into passports.
Above note 77. 80 ATSA, s. 115(c). 81 Ibid., s.115.
See N. Banerjea-Brodeur, ‘Advance passenger information leads to better border control
and faster clearances’, ICAO Journal, Vol. 58, Issue 7, at 30 (September 2002).
S109(3). CAPS was originally developed after the 1996 TWA crash and its revised version
is now being targeted at post-9/11 concerns.

to risk levels, and those who are deemed not to be a risk will likely experience
shorter waiting times at screening points. As mentioned above, IATA’s
‘Simplified Passenger Travel’ initiative also entails screening passengers
at the commencement of their journey and sharing that information
electronically with service providers during subsequent trips. Frequent
travellers posing no risk will experience less delay at screening points as a
The collection of passenger information, whether through PNRs, API,
CAPPS II, SPT or any other means, and the use of that information for
profiling purposes, can ostensibly be an effective tool. Principally, it has the
benefit of reducing wait times for passengers and frequent travellers who are
deemed to be risk-free, and allows screeners to pay more attention to those
thought to pose a threat. In other words, the practice facilitates the optimal
allocation of scarce law enforcement resources to those populations in which
crime is thought to be more likely to occur.
At the same time, however, the extent to which profiling-based methods
optimize the allocation of enforcement resources is heavily dependent on the
quality of available information on ‘risky’ individuals. A biometric system
based on facial or retinal recognition is contingent upon records being
available to facilitate a comparison, and a terrorist with no known previous
record will likely not be detected. Thus, its utility is limited to identifying
known terrorist suspects. In any event, there is still a potential for a false
match rate, given that none of the face, finger or retinal scans are known to be
100 per cent perfect.86 In addition, biometric identification and verification
simply ensure that the person getting on board the plane is a rightful ticket-
holder, and does not prevent would-be terrorists who are properly-ticketed
(and with no known record) to board. Indeed, this method would not have
prevented the 9/11 attacks, as the terrorists responsible did have proper
tickets for those fateful flights.87
In the same vein, the proposed CAPPS II system is fraught with problems –
there are serious concerns over its application, including who operates it and
how it is operated, passengers being incorrectly denied boarding, its hasty
adoption without the benefit of proper economic analysis and reservation
software not being currently able to perform flagging functions. Moreover,
there are concerns over the integrity of the information collected – profile
data can apparently be leaked so that terrorists themselves may be able to use

For more on SPT, see Abeyratne, ‘Attacks on America’, at 83.
See also Daniel, ‘Reform in Airport Security’, at 1635, citing a Department of Defense
study showing that this technology has a higher error rate with regard to false positives.
For a discussion, see Karber, ‘Re-Constructing Global Aviation’, at 798, citing Traci
Watson, ‘US Looks at Which Tech Proposals Will Fly: Government Deluged with Ideas
for Airports, Airliners’ USA Today, at A-4, 26 November 2001.

the profiles as part of their own deceptive strategies.88 For all these reasons,
the TSA has reportedly scrapped CAPPS II, even though it appears that
research is continuing into the development of alternative passenger pre-
screening systems.89
Perhaps the biggest concern over profiling relates to invasion of privacy and
the erosion of fundamental civil liberties.90 Under the US proposal for the
sharing of PNR information, sensitive personal data will be surrendered to
government officials. Such information can be transmitted to other federal
agencies, where it can be collated with independent information collected by
these agencies to build a profile on the relevant passenger. Personal information
such as travel histories and even lodging preferences and meal habits can be
tracked, permitting governments to build up profiles according to racial or
religious grounds. This can open the way to arbitrary discrimination, as certain
individuals of particular ethnicities or religious faiths will invariably be subjected
to greater scrutiny in security operations.91 Indeed, profiling based solely on race
or ethnicity has been a bane which scores of travellers have unpleasantly experi-
enced post-9/11.92 There are also fears that as biometrics-based identification
becomes the entrenched norm, more extensive use will be made of this

R. W. Bloom, ‘Commentary on the Motivational Psychology of Terrorism against
Transportation Systems: Implications for Airline Safety and Transportation Law’ (1998)
25 Transp. L. J. 175, at 179.
See ‘Uncle Sam Mothballs Screening Program’, washingtonpost.com, 16 July 2004, at
See Wong, Chapter 10 in this volume. According to an Associated Press poll taken in
August 2002, two-thirds of American respondents felt concerned that the new security
measures restricted their individual liberties, see Miller, note 4 above, at 408 and Robert
L. Crandall, ‘Security for the Future: Let’s get our airlines flying’ (2002) 67 J. Air L. & Com. 9.
Racial profiling may be permissible under the Fourth Amendment if race is one factor
among several for a decision to stop a person, but not the sole basis. Racial profiling may
also be challenged under the Equal Protection Clause under the Fourteenth Amendment.
Under federal equal protection principles, it may be that racial profiling is generally
prohibited, at least in the absence of finding that it is ‘narrowly tailored’ to serve a
‘compelling state interest’. See generally Samuel G. Gross and Debra Livingston, ‘Racial
Profiling under Attack’ (2002) 102 Colum. L. R. 1413; John Rogers, ‘Note: Bombs, Borders
and Boarding: Combating International Terrorism at United States Airports and the 4th
Amendment’ (1997) 20 Suffolk Transnat’l L. R. 501; and Jamie Rhee, ‘Rational and
Constitutional Approaches’.
For more on racial profiling, see Kareem Shora, ‘Guilty of Flying While Brown’ (2002) 17
Air & Space L. 4; Ellen Baker, ‘Flying While Arab – Racial Profiling and Air Travel Security’
(2002) 67 J. Air L. & Comm. 1375; Charu A. Chandrasekhar, ‘Flying While Brown: Federal
Civil Rights Remedies to Post-9/11 Airline Racial Profiling of South Asians’ (2003) 10
Asian L. J. 215; Ryan L. Bangert, ‘When Airlines Profile Based on Race: Are Claims
Brought Against Airlines Under State Anti-Discrimination Laws Preempted by the
Airline Deregulation Act?’ (2003) 68 J. Air L. & Com. 791; and Richard Sobel, ‘The
Demeaning of Identity and Personhood in National Identification Systems’ (2002) 15
Harv. J. L. & Tech. 319, at 355.

technology beyond security-related purposes. In other words, biometric tech-
nology may slowly desensitize society’s awareness of the right to privacy.
For all these reasons, the EU had been reluctant to comply with US
requirements for PNRs for flights originating in Europe and bound for the
US.93 After extensive negotiations, an agreement was struck in May 2004
between the two sides, providing for the release of information on European
passengers travelling to the US and the strict conditions attached to such
release.94 The conditions restrict which US government agencies can have
access to the information and requires sensitive information such as race and
health to be deleted immediately.95 Overall, the Agreement ensures that the
way in which PNR data is used by US authorities will not unlawfully dis-
criminate against EU passengers and guarantees reciprocal support from the
US for any potential European passenger identification system.
Within the EU, the presence of instruments like Directive 95/46 relating to
the protection of the processing and movement of personal data affords some
level of protection to passengers.96 However, few countries in the world have
such comprehensive laws on privacy protection for their citizens. In
this regard, it would be preferable if a harmonized standard for passenger
information exchange can be instituted through ICAO or IATA auspices,
incorporating the types of safeguards guaranteed by the EU–US Agreement.
Airlines and passengers worldwide may have little choice but to accept that
despite the potential violation of privacy rights, there appear to be few
practical alternatives to data collection and profiling. Without some form
of profiling, the prospect of long delays at airports is a real one, since all

See Communication from the European Commission to the Council and Parliament on
Transfer of Air Passenger Name Records (PNR) Data: A Global EU Approach, Com 2003/
0826 final, 16 December 2003, at 4–5.
See International Agreement on Passenger Name Records between the European
Community and the United States, signed and entered into force 28 May 2004: see
http://europa.eu.int/comm/external_relations/us/news/ip04_694.htm. The Agreement
provides the legal framework under which airlines can transfer PNR data to US authorities
and grants permission to US authorities to access such data held on EU territory.
See Regulation (EC) No. 68/2004 of 15 January 2004 amending Commission Regulation
(EC) No. 622/2003 laying down measures for the implementation of the common basic
standards on aviation security.
1995 O.J.L. 281, 31. Pursuant to this Directive, the EC adopted a so-called ‘adequacy
finding’, based on US undertakings regarding the protection of transferred data. The
Agreement, note 94 above, and the ‘adequacy finding’ are linked, so that the former only
remains in force as long as adequate data protection in the US is granted. The presence of
provisions like Art. 25, para. 6 of Directive 95/46 relating to the protection of the
processing and movement of personal data can be expected to afford some level of
protection, even though Arts. 6, 7 and 13 of the Directive arguably allow for the processing
of data for legitimate purposes and where this is necessary in the public interest, e.g. for the
purpose of aviation security.

travellers would have to be searched systematically. The solution lies in
effecting a balance between security risks and the right of the individual to
his or her privacy. Where exactly this balance falls is, of course, a difficult
question, but any assessment made must take into account the magnitude of
the risks involved, the means of assessing these risks and ultimately, the
weight to be attached to the risks, given the competing imperative of indivi-
dual rights. At the very least, restrictions on the number and type of govern-
ment agencies which can have access to passenger information and the
duration and purpose for which such information should be stored are the
main safeguards which can and must be imposed.

III. Harmonization of aviation security measures
In the light of efforts taken by ICAO and IATA to attain some measure of
harmonization of aviation security measures, care must be taken to ensure
that the measures eventually prescribed are capable of being enforced and
complied with, particularly in the less developed parts of the world. In the
aftermath of 9/11, many of the aviation security initiatives have arisen out of
the developed states like the US and EU countries. Domestically, the US
public’s perception and fear of the risk of a recurrence of terrorist events on
the scale of 9/11 has led to a discernible increase in the government’s will-
ingness to take strong action, particularly on the part of the executive, and
commonly at the expense of civil liberties.
At the same time, it must be recognized that the dynamics of institutional
intercourse in the US and the EU and its implications on civil liberties and the
aviation industry may have important effects internationally. Many of the new
security measures that have been imposed by the US are backed up by sanctions
such as airlines’ exclusion from the US market.97 Thus, foreign airlines wishing
to continue flying to, from and over US airspace have had no choice but to
comply with the regulations. In addition, the US may also ban flights from
foreign airports which do not maintain adequate security standards. This has the
immensely powerful effect of influencing not only the compliance behaviour of
foreign airlines, but of foreign airports and states as well in relation to airport
security measures such as pre-departure passenger and baggage screening. Such
enforcement capability makes up for the inability of bodies like ICAO and IATA
to enforce compliance among states and airlines with prescribed standards.
However, while these measures have the effect of unifying regulations
applicable to US-related flights, it remains doubtful whether such measures,
even if adopted by ICAO for international application, can be properly and
meaningfully implemented worldwide. These doubts are particularly

Paul S. Dempsey, ‘Wide Range of Air Law Instruments Work to Curb Acts of Violence
Against Aviation’, ICAO Journal, Vol. 58, Issue 7, at 30 (September 2003).

pertinent if the measures entail high compliance costs which cannot be easily
borne by other states and their airlines. An example is the proposed use of
biometric technology in passports and other travel documents, which will
entail the use of expensive technology.98 In the developing world, in parti-
cular, it appears unrealistic that such measures will enjoy widespread appli-
cation, at least in the short to medium term.
Hence, ICAO, in its effort to craft aviation security standards that are to
be enforceable globally, must be alive to the fact that a balance will have to
be sought between technologically responsive regulations and realistic expec-
tations for enforceability and compliance in many parts of the world. This
challenge is certainly not new to ICAO – disparate levels of enforcement have
long been observed in relation to matters such as aviation safety, as
mentioned above. Yet, aviation safety has long proceeded on the premise
that the prescription of safety standards worldwide must be uniform, as
encapsulated in ICAO’s SARPs. It is just that the enforcement of the
prescribed SARPs has been replete with problems, not least because it is
impossible for ICAO to demand that every state live up to its obligations
under the SARPs. What this means is that differentiated enforcement may be
inevitable, even if the prescribed standards are fairly uniform. Thus, from a
regime-effectiveness view, it would be desirable if the legislation process can
take into account the realities of differentiated enforcement from the outset,
and the standards prescribed accordingly so as to maximise the chances of
In this regard the aviation industry may not be alone99 in witnessing
the gradual demise of the ‘lowest common denominator’ approach to

Biometric technology is premised upon the identification of passengers using fingerprint,
facial, retinal or iris records.
The industry’s close cousin, the maritime industry, is arguably experiencing a similar
trend, in that states with higher expectations for safety, pollution and security standards
are pressing for high regime standards to be enforced in ports against foreign ships. The
sanctions in such cases would be denied entry or detention of sub-standard ships in ports,
a practice known as ‘port state control’ or ‘port state enforcement’. While such moves can
be pursued unilaterally (in that both legislated standards and enforcement measures in
port can be wholly home-driven), the developed states tend to prefer to ‘up’ the standards
in international conventions adopted by the International Maritime Organization (IMO).
Differentiated enforcement may then ensue, with the developed port states pursuing
stricter enforcement of these measures while developing states are left to grapple with
the complexities of stricter (and usually difficult-to-enforce) regulations. The negative
effects of such a regime could be the flight of sub-standard ships and operators to the
developing world, a prospect that could also happen in the aviation sector. Thus, ‘good’
airlines with commendable safety (and now security) records will continue to fly to the US
and Western Europe, while lesser airlines may concentrate on regional flights in develop-
ing regions of the world where expectations for aviation security are lower. Empirical
proof for such a phenomenon remains, of course, to be researched.

international regime formation, one characterized by the tendency to pre-
scribe standards that can ideally be met by the weakest of participants in the
decision-making process. By the same token, neither is it desirable to pre-
scribe ‘maximal’ standards that are so stringent that they do not afford
meaningful compliance by weaker target actors. Thus, it may be that a
more differentiated ‘two-track’ approach will have to be adopted – an
approach consisting of, on the one hand, minimal ‘baseline’ standards
applicable to all states, and beyond that, the flexibility to recognize the need
for higher standards by certain interested states.
In this regard, differentiation in measures may be contingent upon at least
two variables – the subject matter of the standard itself, and its geographical
application. Both variables are, in turn, linked to the difference in perception
of risks by different actors and the weight attached to such risks. Thus, a
particularly costly measure (such as biometric technology or air marshals)
may be less appealing to a state in a region of the world which sees little reason
to expend huge resources on what may appear to be a minimal terrorist risk.
Yet, at the same time, the state may be persuaded as to the cost-effectiveness
of a fortified cockpit door or increased baggage surveillance.
In this regard, the individual aviation security measures discussed in this
chapter should not be viewed in isolation, but as a package of numerous
alternatives that can be tailored to meet the specific demands and needs of
states. In other words, ICAO should afford the possibility of states using a
combination of measures that best reflects their specialized needs, rather than
imposing fixed standards on all states. The compliance of states with these
measures can then be taken up either by ICAO’s auditing scheme, or by
individual states requiring compliance as a condition of entry into their
airspace. This flexible approach will maximize the chances of compliance
by states with the requisite standards.

IV. Conclusion
The broad themes dealt with in this chapter revolve around the strengths and
weaknesses of aviation security measures proposed and implemented, and the
best possible means to harmonize these measures given the differentiated
perception of risks in various countries. On the latter issue, the complete
harmonization of universal standards is counter-productive, as this will only
lead to chronic non-compliance among many countries. Instead, interna-
tional agencies like ICAO should seek to impose a certain band of minimal
standards applicable to all actors, with the allowance for some to impose
more stringent standards, even if these had to be unilaterally enforced.
Following from this, and relating it to the first issue, it can be appreciated
that while armed marshals and pilots can act as a supplementary line of

defence against terrorism,100 the risks of having guns on board planes may be
too high for some states to accept. This is particularly so if these states view
themselves and their airlines to be less at risk of terrorist attacks. Fortifying
cockpit doors represents a more viable solution as they entail less costs and
minimal safety risks. Thus, flight sectors which are viewed to be less terrorist-
prone should be allowed to settle for a less expensive alternative such as
fortified cockpit doors, as opposed to air marshals. In other words, the
different aviation security measures should be viewed as a coherent whole
in tailoring the response of states, with flexible combinations of such mea-
sures being prescribed for states with different needs.
As for baggage screening, the benefits of stopping dangerous items from
being brought onto the plane are clear, for flights anywhere in the world. In
this respect, the solution appears to be for countries to work together to
develop more cost-effective technologies for baggage screening. On their
part, passenger screening initiatives come at high financial as well as privacy
infringement costs.101 For the sake of protecting privacy, a balance must be
found between the amount of information required from passengers and the
use to which government agencies put such information. Privacy protection
should ideally be extended to the citizens of all countries, not only those
which have the laws to ensure this.
Of course, it remains a worthy ideal to have all states and airlines pursue all
security measures within a harmonized regime, regardless of the particular
level of risk that exists for different countries. Such harmonization is parti-
cularly desirable for a global industry such as aviation, where the ‘weakest
link’ in a particular actor may have ramifications for other states. However,
the realities are such that stringent standards will be viewed as necessary by
some, but not all states, and a level of appreciation must be given to the
different perceptions of risks by different actors. Ultimately, the overriding
challenge remains for the international community to develop a regulatory

A concept known as ‘layering’ in the US whereby the hope is for ground security
measures to track down any security breaches so that onboard measures are put into
use only on the rarest of occasions as a last line of defence.
In terms of cost, the US government is reportedly seeking US$435 million more from the
airline industry for passenger screening, arguing that the airlines should be paying for
their share of aviation security. IATA views this as a new tax and is vigorously opposing it,
pointing out that carriers already gave the government US$2 billion every year to prevent
terrorist attacks. IATA also argues that financially struggling airlines cannot pass new
taxes onto passengers because of intense competition. Air carriers now administer a $2.50
security tax onto passenger tickets, which amounts to about $1.6 billion annually.
Airlines are also assessed $315 million every year for passenger screening, a responsibility
the federal government assumed soon after the 9/11 attacks. The airlines were given a
four-month reprieve last year from both the passenger security tax and the annual
assessment because the war in Iraq was expected to hurt their business, see the Straits
Times (Singapore), 1 May 2004.

regime that is not only highly effective in preventing acts of aerial terrorism,
but one which is responsive to the different security needs of different actors,
and which does not, in the overall scheme, unduly interfere with the effi-
ciency and productivity of commercial aviation, impose excessive costs,
create unwarranted passenger inconvenience or intrude unnecessarily into
individual privacy or civil liberty.102

Dempsey, ‘Aviation Security’, at 721–722.

International responses to combat maritime terrorism

I. Introduction
The terrorist attacks in the United States on 9/11 shocked not only the United
States, but the entire international community. The attacks were unequivocally
condemned by the United Nations Security Council and by most members of the
international community. They triggered an almost immediate response led by
the United States at the international level for additional measures and increased
cooperation to prevent and suppress terrorist activities.
As a result of 9/11, states and international organizations were forced to
completely rethink the threat of maritime terrorism. They recognized that if
terrorists groups could strike powerful states using commercial aircraft, they
could also strike using commercial shipping. The threat of maritime terror-
ism suddenly included the following: oil tankers being hijacked and used as
weapons against other ships or port facilities; terrorists entering countries
posing as seafarers, and weapons of mass destruction being shipped on
merchant ships to terrorist organizations.
The United States recognized that the threat of maritime terrorism could
not be dealt with unilaterally. International shipping is by its very nature,
international, and can only be regulated through international cooperation.
Since 9/11 the United States has led a two-pronged approach to obtain
international cooperation to deal with the threat of maritime terrorism.
First, it has worked vigorously and patiently to encourage the relevant
United Nations bodies such as the UN Security Council, the International
Maritime Organization (IMO) and the International Labour Organization
(ILO) to take action that requires member states to impose new measures to
deal with the threat of maritime terrorism. Second, the United States has used
its power and influence to get cooperation from other states through bilateral
agreements and the establishment of ‘coalitions of the willing’. As a result,
established principles of international law have been adapted and changed,
and in some cases, challenged.
In this chapter, I will outline the actions and measures that have been
taken and are being taken to prevent and suppress acts of terrorism against
international shipping. I will discuss how the actions and measures have

presented challenges to existing principles of international law, and evaluate
how principles of international law have been adapted or modified to meet
the challenges posed by maritime terrorism.

II. Actions to enhance port and ship security
A. IMO actions to enhance port and ship security
Following the events of 9/11, the United States initiated and led the drive
at the IMO to adopt measures to strengthen maritime security on ships and in
ports.1 As a result, the IMO undertook a thorough review of measures
to combat acts of violence and crime at sea. At the 22nd Assembly meeting
in November 2001, it was agreed to hold a Conference on Maritime Security
in December 2002 to adopt new regulations to enhance ship and port
After a year of preparations the Conference of Contracting Governments
to the International Convention for the Safety of Life at Sea 1974 was held
from 9 to 13 December 2002. The Conference adopted resolutions containing
a series of measures to strengthen maritime security and prevent and suppress
acts of terrorism against international shipping. Many of the measures were
adopted through amendments to the major IMO Convention governing
the safety of ships, the 1974 International Convention for the Safety of Life
at Sea (SOLAS).3 The 2002 amendments to SOLAS entered into force on
1 July 2004.

1. Actions in December 2002 to enhance maritime security
One of the most important measures taken at the 2002 Conference was to add
a new Chapter on maritime security to SOLAS. The new Chapter XI-2 is

On 15 January 2002 the United States submitted a proposal to the 75th Session of the IMO
Maritime Security Committee on measures to improve maritime security (IMO Doc MSC
75/ISWG/5/7) (text provided to author). The proposal covered the following areas:
Automatic Identification Systems, Ship and Offshore Facility Security Plans, Port Facility
Security Plans, Ship Security Officers, Company Security Officers, Seafarer Identification
Verification and Background Check, Port Vulnerability Assessments, Port of Origin
Container Examinations, Cooperation with the World Customs Organization,
Information on the Ship and its Cargo and People, Means of Ship Alerting, and Ship
Security Equipment.
For a summary of the actions of the IMO on maritime security, see http://www.imo.org.
For a summary of the measures adopted in December 2002, see http://www.imo.org/
Newsroom/mainframe.asp?topic_id ÂĽ 583 &doc_id ÂĽ 2689. The December 2002 amend-
ments to the SOLAS convention are available online in the Australian Treaty Series (not yet
in force), [2003] ATNIF 11, http://www.austlii.edu.au/au/other/dfat/treaties/notinforce/

entitled ‘Special Measures to Enhance Maritime Security’. Among the special
measures included in Chapter XI-2 are the following.
First, flag states are required to set security levels for their ships, and port
states are required to set security levels for their port facilities. Ships entering
a port or in port are required to comply with the security level of the flag state
or port state, whichever is higher. Second, the master of a ship is required to
have information on board the ship concerning persons or organizations
responsible for the employment of crew members of the ship. Third, ships
constructed after 1 July 2004 are required to be provided with a Ship Security
Alert System. The deadline for installing a Ship Security Alert System on ships
constructed before that date depends upon the type of vessel. Fourth, it is
provided that the master of a ship has the overriding authority and the
responsibility to make decisions and measures with respect to the safety
and security of the ship.

2. Adoption of ISPS Code
Supplementary to Chapter XI-2 is the International Ship and Port Facility
Security (ISPS) Code.4 The 2002 Amendments to SOLAS provide that ship-
ping companies and ships must comply with the requirements of the ISPS
Code. As the title suggests, the ISPS Code contains measures designed to
enhance the security of ships and the security of port facilities. The ISPS Code
has two parts, A and B. Part A is mandatory for the purpose of compliance
with Chapter XI-2. Part B is to be used as a guide and treated as recommen-
datory. The following requirements are prescribed in the ISPS Code.
First, a ship is required to carry on board a Ship Security Plan approved by
the flag states on the basis of a Ship Security Assessment. Second, a company
operating a ship must designate a Company Security Officer (CSO) for every
ship, and every ship is required to have a designated Ship Security Officer
(SSO). The CSO and the SSO are required to undergo training in maritime
security in accordance with the guidance given in Part B of the ISPS Code.
Third, drills and exercises with respect to the Ship Security Plan are required
to be carried out at appropriate intervals by all parties concerned. The CSO
and appropriate shore-based personnel are also required to participate in the
drills and exercises to ensure effective shore-ship coordination with respect to
the Ship Security Plan. Fourth, a ship will be issued an International Ship
Security Certificate (ISS Certificate) after verification that the ship complies
with Chapter XI-2 and the ISPS Code. Fifth, a ship is required to act upon
the security levels set by the port state or the flag state by carrying out
the activities prescribed in the ISPS Code with the aim of identifying and
taking preventive measures against security incidents, which are defined in

The full text of the ISPS Code is available online in the Australian Treaty Series with the
other amendments adopted in December 2002. Ibid.

Chapter XI-2 as ‘any suspicious act or circumstance threatening the security
of the ship’. Sixth, Port Facilities to which Chapter XI-2 applies are required
to develop and maintain a Port Facility Security Plan on the basis of a Port
Facility Security Assessment. These facilities are also required to designate
Port Facility Security Officers who, together with appropriate port facility
security personnel, are required to undergo training in maritime security in
accordance with the guidance given in Part B of the ISPS Code. They are also
required to conduct drills and exercises with respect to the Port Facility
Security Plan.

3. Compliance and control measures by port states
Ships are subjected to Port State Control with respect to compliance with
Chapter XI-2. The Port State Control inspection is limited to verifying that
there is on board a valid ISS Certificate issued under the provisions of Part A
of the ISPS Code. When a valid ISS Certificate cannot be produced or when
there are clear grounds for believing that a ship is not in compliance with the
requirements of Chapter XI-2 or Part A of the ISPS Code, certain control
measures may be taken by the port state against the ship. Such control
measures include inspection of the ship, delaying the ship, detention of the
ship, restriction of operations including movement within the port, or
expulsion of the ship from port.
In addition, a port state may require that ships provide information to
ensure compliance with Chapter XI-2 prior to entry into port, including
information relating to the ISS Certificate, the security level of the ship, the
security level at previous port calls, and security measures taken at previous
port calls.

4. Other measures to enhance maritime security on ships
The 2002 Amendments to SOLAS also contain other measures to enhance
maritime security, including the following.
First, they brought forward the dates by which certain ships had to install
an Automatic Identification System (AIS). The AIS enables shore facilities to
automatically identify ships and obtain basic information about them.
Second, the regulations require that a ship’s Identification Number must be
permanently marked in two places on the ship, one of which must be clearly
visible. Third, ships are required to carry on board a Continuous Synopsis
Record, which is intended to provide an on-board record of the history of the
ship with respect to the information recorded therein.

5. Significance of the new maritime security measures
The new maritime security measures did not pose any challenge to existing
rules of international law. They gave new responsibilities to port states and
imposed new obligations on flag states and shipping companies, but these

measures are consistent with the provisions of international law as set out in
the 1982 UN Convention on the Law of the Sea (UNCLOS).
The measures adopted by the IMO were very significant in two respects.
First, they expanded the IMO’s traditional responsibility for maritime safety
to include maritime security. Maritime security suddenly became an integral
part of the responsibilities of the IMO. Second, they expanded the IMO’s
rule-making authority into port facilities, an area that had previously been
considered a matter within the domestic jurisdiction of the port states
because port facilities are within their territorial sovereignty.
The new measures required flag states, port states and shipping companies
to make maritime security a major priority so that they would be able to meet
the 1 July 2004 deadline for entry into force of the new measures. Some
developing countries have had a very difficult time attempting to meet the
deadline. The result, however, is that in most areas of the world port facilities
and ships became more secure after 1 July 2004.
The new measures can also be viewed as a major victory for the United
States in the field of multilateral diplomacy. The measures finally adopted by
the IMO in December 2002 are very similar to those made by the United
States in its proposal to the IMO in January 2002.5 The United States used its
power and influence to get almost the entire international community to
agree to adopt and implement comprehensive new measures on ship security
and port security, notwithstanding the fact that many might not have had the
same level of concern about maritime terrorism as the United States.

B. ILO actions on seafarers’ identity documents
One of the other major concerns of the United States was that there was no
reliable international system for identifying seafarers who enter the United
States as crew members of a ship. It was generally known that many seafarers
carried false documents and that the identity systems used for seafarers in
many developing countries were not secure. This was one of the issues that
had been included in the United States proposal to the IMO Maritime
Security Committee in January 2002,6 but it was later decided that this matter
should be dealt with by the International Labour Organization (ILO) rather
than the IMO.
On 3 June 2003, at the ninety-first session of the governing body of the
ILO, major revisions were made to the Seafarers’ Identity Documents
Convention of 1958.7 The revisions were intended to address the continuing
threat posed by maritime terrorism to the security of passengers and crews

US Proposal, above note 1. 6 Ibid.
The text of the Revised Convention is available at: http://www.ilo.org/ilolex/cgi-lex/

and to the national interests of states, in a manner that was consistent with the
mandate of the ILO to promote decent conditions of work for seafarers.
The Seafarers’ Identity Documents Convention (Revised) 2003 will come
into force six months after the date on which the ratifications of two
Members have been registered with the Director-General of the ILO.8 It
provides that each Member shall issue to each of its seafarer nationals a
seafarer’s identity document that is designed to prevent tampering and
falsification while at the same time being generally accessible to governments
at the lowest possible cost. The identity document must include at least one
security feature such as watermarks, ultraviolet features, holograms or laser
engraving. Members are also required to keep a record of identity documents
in an electronic database, and to provide a focal point for responding to
inquiries from the authorities of other Members. The Convention is generally
seen as a move that will further improve security in the global shipping
Following its adoption in June 2003 there were negotiations and discus-
sions on the technical features and on the economic and political implica-
tions of the new convention. On 26 March 2004 the ILO adopted a new
‘biometric’ identity verification system which sets a new global standard. It
allows for the use of a ‘biometric template’ for turning two seafarer’s finger-
prints into an internationally standardized barcode on the Seafarer’s Identify
Document. The new global standard will enable biometric identification of
the world’s seafarers.9
Most of the 1.2 million seafarers who handle 90 per cent of the world’s
trade are from developing countries, and it will be difficult for them to meet
the standards for the new identify document. For example, the Philippines is
one of the world’s leading suppliers of the world’s seafarers, and more than
half of the 500,000 Filipino seafarers are officially employed overseas. The US
Trade and Development Agency (USTDA) has awarded a grant of
US$515,000 to the Philippines government to fund technical assistance
related to the development of a seafarers’ identity system that would meet
the requirements of the 2003 Convention.10
The developments on seafarers’ identity documents are another example
of how the international community moved with unprecedented speed to
modernize its rules and standards in response to the threat of maritime

The Revised Convention will come into force from 9 February 2005. The go-ahead follows
ratification of the convention by Jordan and France, fulfilling the entry into force
requirement of two member states supporting it.
ILO Press Release, Friday 26 March 2004 (ILO/04/12).
TDA Press Release, 30 September 2003, http://www.tda.gov/trade/press/Sept30_03.html.

C. US Customs Container Security Initiative (CSI)
At the same time as the United States initiated actions at the IMO and ILO to
enhance maritime security, it also initiated actions outside of existing inter-
national institutions to deal with the threat of maritime terrorism. The first of
its initiatives to establish coalitions of willing partners to cooperate to combat
the threat of maritime terrorism was the Container Security Initiative (CSI).
In January 2002 the US Customs proposed the CSI in order to secure what
it determined to be an indispensable, but vulnerable link in the chain of global
trade: the sea container. The objective of the CSI is to prevent global contain-
erized cargo from being exploited by terrorists. The fear was that sea contain-
ers could be used by terrorists to bring weapons of mass destruction (WMD),
or even terrorists themselves, directly into ports in the United States.
The United States maintained that ensuring the security of the maritime
trade system is essential because approximately 90 per cent of the world’s
cargo moves by container. Some 200 million sea cargo containers move
annually among the world’s top seaports, and nearly 50 per cent of the total
value of all US imports arrive via sea. In 2001 US Customs processed more
than 214,000 vessels and 5.7 million sea containers.11
The objective of the CSI is to first engage the ports that send the highest
volumes of container traffic into the United States, as well as the governments
in these locations, in a way that will facilitate detection of potential problems
at the earliest possible opportunity. Since approximately two-thirds of all US
bound sea containers pass through twenty major seaports around the globe,
the United States initially focused on these twenty ports. As of August 2004,
nineteen of the twenty ports have committed to joining CSI and are at various
stages of implementation, and ports in several other countries have also
joined the initiative.12
CSI consists of four core elements:
using intelligence and automated information to identify and target contain-

ers that pose a high risk;
 using advanced detection technology to quickly pre-screen containers that
pose a risk;
 using smarter, more secure containers with tamper-proof seals which alert
authorities to tampering while the container is in transit;
 pre-screening those containers that pose a risk, at the port of departure
before they arrive at US ports.

US Customs Press Release, 22 February 2002, http://www.cbp.gov/xp/cgov/newsroom/
CSI in Brief, http://www.customs.gov/xp/cgov/enforcement/international_activities/csi/.
This site gives detailed information on the CSI, including a list of ports, a fact sheet, etc.

Under the CSI programme a small number of officers from the US
Customs and Border Protection office are deployed in foreign ports to
work together with their host nation counterparts to pre-screen containers
that will be shipped to the United States. The US officers assist in identifying
high risk containers that will be subject to screening, but the actual screening
or inspection is done by officers from the host country. The programme is
reciprocal, and the United States offers participating countries the right to
send their customs officers to major ports in the United States to target
ocean-going cargo to be sent by sea container to their countries. Japan and
Canada currently station their customs personnel in United States ports
under this reciprocal arrangement.
The CSI is an example of an initiative by the United States to establish a
framework for international cooperation outside of existing international
institutions and international regulations. The United States used its bargain-
ing power to persuade countries to cooperate. Given the threat of delays for
ships coming to the United States from ports that were not part of the CSI,
states decided that it would be in their economic interests to cooperate with
the United States. Some may even have felt that it would give their ports a
competitive economic advantage over rival ports that were not part of the
CSI. Therefore, states were willing to accept what before 9/11 might have been
regarded as an infringement of their sovereignty.
Critics of United States unilateralism can argue that the CSI is directed at
protecting the United States and is pushing the risk to foreign ports. They can
argue that a better approach might have been to work through established
international bodies such as the IMO and International Customs
Organization (ICO) to establish a system for securing containers which
protects all countries, and that the CSI might undermine efforts to develop
international regulations and standards through the IMO or the ICO.13
Nevertheless, given the fact that most major ports have joined the initiative,
one must conclude that it has been a success.

III. Actions to interdict suspect vessels at sea
A. Proliferation Security Initiative (PSI)
The Proliferation Security Initiative (PSI) is an initiative of the United States
to establish a coalition of willing partners to respond to the growing challenge
posed by the proliferation of WMD. Like the CSI, it is an attempt by the

For a critical analysis from an Australian perspective, see Nigel Brew, ‘Ripples from 9/11:
the US Container Security Initiative and its Implications for Australia’, Information,
Analysis and Advice for the Parliament, 13 May 2003, at http://www.aph.gov.au/library/

United States to create a framework for international cooperation outside the
international organizations and international treaties.
The PSI was announced by United States President George Bush in Poland
on 31 May 2003, just prior to the G8 Summit. President Bush stated that the
United States and its close allies would search planes and ships carrying
suspect cargo and seize illegal weapons or missile technologies.14 As the PSI
has developed, it has been based primarily upon the development of coop-
erative arrangements among a coalition of participating states regarding the
interdiction of ships suspected of carrying WMD and missile-related
The United States began working with ten other countries in 2003 to
develop a set of principles that would identify practical steps to interdict
shipments of WMD, their delivery systems or related materials flowing to or
from ‘state or non-state actors of proliferation concern’. The ten countries
were Australia, France, Germany, Italy, Japan, the Netherlands, Poland,
Portugal, Spain and the United Kingdom.15 At a meeting in Brisbane in
September 2003 the ten original participating countries agreed that North
Korea and Iran were of particular proliferation concern. At that meeting the
participating states also agreed to a ‘Statement of Interdiction Principles’.16
The ten original participating countries were later joined by five new parti-
cipating countries – Canada, Denmark, Norway, Singapore and Turkey. At
the first anniversary meeting of the PSI in May 2004 it was stated that more
than 60 states had expressed support for the PSI, and that Russia had decided
to join the core group of PSI participating states.17

1. PSI interdiction principles and the law of the sea
In the Statement of Interdiction Principles participating states agreed to take
certain specific actions in support of interdiction efforts regarding cargoes of
WMD, their delivery systems or related materials, to the extent their national
legal authorities permit and consistent with their obligations under interna-
tional law and frameworks. Questions have been raised as to whether some of
the actions called for in the Statement of Interdiction Principles are consis-
tent with existing rules of international law. Of particular concern was
whether the principles with respect to interdiction at sea are consistent with

Remarks by the President to the People of Poland, Wawel Royal Castle, Krakow, Poland,
See US State Dept web page http://www.state.gov/t/np/c10390.htm.
For the full text, see http://www.state.gov/t/np/rls/fs/23764.htm.
Undersecretary John R. Boulton, Press Conference on the Proliferation Security Initiative,
31 May 2004 http://www.state.gov/t/us/rm/33556.htm.

the provisions of UNCLOS. This concern is especially relevant because the
United States is not a party to UNCLOS.18
In May 2004 the Bureau of Nonproliferation of the US State Department
published ‘Frequently Asked Questions’ on the PSI on the Internet.19 On
28 July 2004 the United States published a new brochure on PSI on the
Internet which is intended to answer basic questions about the PSI.20 These
documents seem to be intended to help states understand the PSI and to
alleviate their concerns. For example, one of the issues that has been of
concern was whether the PSI would affect legitimate commerce in dual use
equipment. The Frequently Asked Questions address this issue as follows:
PSI is not aimed against legitimate commerce, dual-use or otherwise . . .
PSI does not envision stopping and inspecting every shipment that might
involve items that could be used in a WMD- or missile-related proliferation
program; rather the United States intends to take action based on solid
information. Legitimate dual-use commerce will very rarely be affected
by PSI.
In the Interdiction Principles states agree, on their own initiative, to board
and search any suspect vessels flying their flag in their internal waters or
territorial seas or in areas beyond the territorial seas of any state. This is
consistent with the principles governing the law of the sea. States have a right
to board and search vessels in their internal waters and territorial sea which
are flying their flag because the sovereignty of a state extends to its internal
waters and to its territorial sea and a state’s laws apply on ships flying
their flag.21
A major concern with PSI was the circumstances under which the inter-
diction principles were intended to allow participating states to board and
search foreign vessels in international waters.22 Under UNCLOS, high seas
principles govern jurisdiction over ships in such areas. The general principle
is that ships on the high seas are subject to the exclusive jurisdiction of the
flag state, except in exceptional circumstances provided for in UNCLOS or
other international treaties.23

As of 16 July 2004, there were 145 parties to UNCLOS, but the United States was not one
of them. See http://www.un.org/Depts/los/.
See http://www.state.gov/t/np/rls/fs/32725.htm.
See http://www.state.gov/t/np/rls/other/34726.htm.
UNCLOS, Articles 2 and 92.
‘International waters’ is not a term of art in the law of the sea or in UNCLOS. However, it
generally refers to waters that are seaward of the territorial sea of any state, that is, outside
the territorial sovereignty of any state. Thus international waters would include the
exclusive economic zone as well as the high seas because under Article 58(2) of
UNCLOS, the principles of jurisdiction governing vessels on the High Seas also apply in
the exclusive economic zone.
UNCLOS , Article 92.

If the interdiction principles are interpreted in a manner that is consistent
with the principle that ships in international waters are subject to the exclu-
sive jurisdiction of the flag state, the interdiction of ships in international
waters must be premised on the principle of flag state consent. Participating
states would have a right to board and search vessels flying the flag of another
state only if the flag state consents to such boarding and search. The inter-
diction principles provide that a participating state should seriously consider
giving other states consent to board and search vessels flying its flag under
appropriate circumstances. This clause is consistent with the principle of flag
state consent. Among states participating in the PSI it would be expected that,
if one PSI state were to request permission to board and search a ship flying
the flag state of another PSI state, the latter PSI state would favourably
consider the request. The Frequently Asked Questions published in May
2004 make it clear that interdiction of ships in international waters will be
premised on the consent of the flag state.24
The interdiction principles also provide that participating states should
take action regarding vessels in their ports and internal waters as well as
vessels entering or leaving their ports or internal waters. Participating states
agree to stop and/or search vessels in their ports or internal waters when such
vessels are reasonably suspected of carrying such cargoes to or from states or
non-state actors of proliferation concern. Participating states also agree to
enforce conditions on suspect vessels entering or leaving their ports and
internal waters. Such conditions might include a requirement that such
vessels be subject to boarding and search prior to entry. Although there is
no provision in UNCLOS directly on this point, such actions would be
consistent with principles of international law. Ports and internal waters are
within the territorial sovereignty of a state, and states may impose conditions
on vessels in its ports and internal waters and on vessels in its territorial sea
that intend to enter its ports or internal waters.
The most controversial actions set out in the interdiction principles con-
cern actions of coastal states with regard to vessels in their territorial sea or
contiguous zone, and with regard to vessels entering or leaving their territorial
sea. These are controversial because under UNCLOS the vessels of all states
have a right of innocent passage through the territorial sea of all states.25 Passage
is innocent so long as it is not prejudicial to the peace, good order or security of

Note 19 above. One of the answers states that: ‘PSI actions will be taken consistent with
existing national legal authority and international law and frameworks. This includes
relevant international legal principles relating to boarding of vessels on the high seas. In
the case of interdiction of vessels flying flags of convenience, the consent of the flag
state would ordinarily provide a clear basis for a boarding on the high seas under inter-
national law’.
UNCLOS, Articles 17–19.

the coastal state. It is difficult to argue that the mere passage of a vessel
containing component parts for making weapons of mass destruction through
the territorial sea of a coastal state is prejudicial to its peace, good order or
Furthermore, special passage rules for the vessels of all states apply in
straits used for international navigation which fall within the territorial sea
of the littoral states. The vessels of all states have the right of transit passage
through straits used for international navigation, and such right cannot be
impeded or suspended by the littoral states.26 The right of transit passage is an
even broader right than the right of innocent passage.
Therefore, states parties to UNCLOS are not likely to interfere with the
vessels exercising the right of transit passage through a strait used for inter-
national navigation or the right of innocent passage through the territorial sea.
Instead, they are likely to take the position that any action taken with respect
to such vessels must be consistent with their obligations under international
law as set out in UNCLOS. However, it would be legal for the coastal state to
board and search a suspect vessel passing through its territorial sea if the flag
state of the suspect vessel expressly authorized or requested such action by
the coastal state.
There is one other circumstance in which it may be legal for a coastal state
to board and search a suspect vessel in its contiguous zone or territorial sea. If
a suspect vessel was on a route which indicated that it was intending to bring
WMD into the territory of the coastal state in violation of its customs laws
and regulations, such vessel would not have a right of innocent passage. It
could be boarded and searched by the authorities of the coastal state in its
territorial sea. Similarly, it could be boarded and searched in the contiguous
zone, which is a zone adjacent to the territorial sea in which the coastal state
has special powers to enforce certain of its domestic laws, including its
customs laws.27
Unfortunately, the Frequently Asked Questions issued by the State
Department do not address the difficult issues relating to interdiction by
coastal states of vessels exercising the right of transit passage or the right of
innocent passage.

2. PSI and the international legal system
The United States has argued that the PSI builds on efforts by the interna-
tional community in existing treaties and regimes to prevent the proliferation
of WMD. It has pointed out that the PSI is consistent with and a step which
implements the Statement of the President of the UN Security Council in
January 1992, when he stated that the proliferation of all WMD constitutes a

26 27
UNCLOS, Article 38. UNCLOS, Article 33.

threat to international peace and security. Supporters of the PSI argue that
the PSI is justified because of the threat and danger posed should WMD fall
into the hands of rogue states or terrorist organizations.28
Critics of the United States can argue that the PSI is another example of the
United States using its power to establish an international framework for
creating and enforcing norms outside the framework of existing international
conventions and international institutions. They point out that the existing
legal regimes governing nuclear non-proliferation do not contain any provi-
sions authorizing member states of the UN to enforce the principles of non-
proliferation through the interdiction of vessels carrying WMD. They argue
that the PSI is a threat to the existing international legal order because the
United States and its partners are not acting within the framework of the
relevant international institutions to build upon and develop existing inter-
national legal regimes to meet new problems. Instead, the PSI is an example of
the United States using its influence to create a ‘coalition of the willing’ to act
outside of existing international legal regimes and conventions.29

B. Bilateral ship-boarding agreements
One of the goals of the United States under the PSI is to enter into bilateral
agreements with major flag states which give the United States permission to
board and search vessels flying their flag when such vessels are suspected of
carrying WMD.
The first such boarding agreement was signed by the United States and
Liberia on 11 February 2004.30 The boarding agreement provides the United
States with authority to board vessels flying the flag of Liberia if they are
suspected of carrying illicit shipments of WMD. Liberia has the world’s
second largest ship registry. According to the United States, this boarding
agreement is an important step in further operationalizing the PSI and
strengthening the mechanisms that the United States has at its disposal to
interdict suspect WMD-related cargoes.
On 12 May 2004 the United States and Panama signed the bilateral Panama –
United States PSI Ship-Boarding Agreement.31 Panama is the world’s largest ship
registry. Now that the United States has bilateral ship-boarding agreements with
the world’s two largest ship registries in order to supplement the boarding

Baker Spring, ‘Harnessing the Power of Nations for Arms Control: The Proliferation
Security Initiative and Coalitions of the Willing’, Backgrounder, No. 1737, The Heritage
Foundation, 18 March 2004.
Articles critical of the PSI include ‘The Proliferation Security Initiative: The Legal
Challenge’, Bipartisan Security Group Policy Brief, September 2003, www.gsinstitute.
For a summary and the full text, see http://www.state.gov/t/np/c12387.htm.
For a summary and the full text, see http://www.state.gov/t/np/c12390.htm.

arrangements it has with its coalition partners, almost 50 per cent of the total
commercial ships in dead weight tonnage is subject to the rapid consent proce-
dures for boarding, search and seizure by United States forces.32
The bilateral ship boarding agreements are consistent with UNCLOS. They
are premised on the principle of flag state consent to the boarding of ships
in international waters. They contain a procedure for obtaining expedited
consent of the flag state to a request for boarding from the United States if
there are reasonable grounds to suspect that a ship of the flag state is engaged
in proliferation by sea. They also contain a ‘tacit authorization’ provision
which provides that if the authorities in the flag state do not respond within
two hours to a request from law enforcement officers of the other state to take
appropriate measures, the law enforcement officers may proceed to board
and search the suspect ship.
The ship-boarding agreements complement the special arrangements the
United States has with its partners under the PSI. If the United States has
arrangements containing procedures for expedited consent with its PSI
partners as well as the major flag states that are not members of the PSI, it
will have procedures in place which will enable it to seek the consent of flag
states to board a significant percentage of the world’s tonnage of vessels.

C. Proposed protocol to 1988 SUA Convention
1. The SUA Convention
The global convention which addresses the issue of maritime terrorism is
the 1988 Convention for the Suppression of Unlawful Acts Against the Safety
of Maritime Navigation (SUA Convention).33 The SUA Convention was
adopted in Rome on 10 March 1988 and entered into force on 1 March
1992. The IMO serves as the secretariat and depository for the Convention.
The SUA Convention and Protocol follow the scheme that was first
established in the 1970 Convention for the Suppression of Unlawful Seizure
of Aircraft.34 The scheme of the 1970 Hijacking Convention has been fol-
lowed in several other conventions that are commonly referred to as the
‘terrorist conventions’.35 The scheme in all of the conventions is to establish
‘universal jurisdiction’ among states parties to the Convention.

US State Department, Daily Press Briefing, 12 May 2004, http://www.state.gov/r/pa/prs/
The text of the SUA Convention is available on the home page of the Australia Treaties
Library at: http://www.austlii.edu.au/au/other/dfat/treaties/1993/10.html.
The text of the 1970 Hijacking Convention is available on the Australia Treaties Library
website at: http://www.austlii.edu.au/au/other/dfat/treaties/1972/16.html.
The other ‘terrorist conventions’ following the scheme of the 1970 Hague Convention
include: (1) Convention for the Suppression of Unlawful Acts against the Safety of Civil
Aviation, signed at Montreal on 23 September 1971; (2) Convention on the Prevention

The SUA Convention applies to the following acts that endanger the safety
of international maritime navigation:
seizure of or exercise of control over a ship by any form of intimidation;

violence against a person on board a ship;

destruction of a ship or the causing of damage to a ship or to its cargo;

placement on a ship of a device or substance which is likely to destroy or

cause damage to that ship or its cargo;
destruction of, serious damaging of, or interference with maritime naviga-

tional facilities;
knowing communication of false information;

injury to or murder of any person in connection with any of the preced-

ing acts.
States parties have an obligation to make the above offences a crime under their
laws when the alleged offence takes place in their territory. In addition, all states
parties must establish jurisdiction over the offence when the alleged offender is
‘present in their territory’, even though it has no other links to the alleged offence.
States parties also have an obligation to take alleged offenders into custody
if they enter their territory, and to either extradite them to another state party
that has jurisdiction, or to turn the case over to their own authorities for
prosecution in their courts. This is generally referred to as the obligation to
‘extradite or prosecute’. The obligation of a state to arrest alleged offenders
who enter their territory applies no matter where the offence took place.

2. United States proposal for a new protocol
In October 2001, one month after 9/11, the Legal Committee of the IMO
decided to review the SUA Convention in the wake of the terrorist attack on
the United States. The Legal Committee agreed to include the review of the
SUA Convention as a priority item in its work programme.36
In April 2002 the Legal Committee agreed to establish a Correspondence
Group led by the United States with the short-term aim of developing a working
paper on the scope of possible amendments for consideration at the 85th session
of the Committee. The longer aim was to draft the amendments and make a
recommendation to the IMO Assembly that it convene an international
and Punishment of Crimes against Internationally Protected Persons, including
Diplomatic Agents, adopted by the General Assembly of the United Nations on
14 December 1973; (3) International Convention against the Taking of Hostages, adopted
by the General Assembly of the United Nations on 17 December 1979; (4) International
Convention for the Suppression of Terrorist Bombings, adopted by the General Assembly
of the United Nations on 15 December 1997; (5) International Convention for the
Suppression of the Financing of Terrorism, adopted by the General Assembly of the
United Nations on 9 December 1999.
IMO Legal Committee, 83rd Session, 8–12 October 2001. A summary of the work of the
Legal Committee is available at www.imo.org.


ńňđ. 9
(âńĺăî 23)