The Wrongs of Unlawful Immigration

Abstract

For too long, criminal law scholars overlooked immigration-based offences. Claims that these offences are not ‘true crimes’ or are a ‘mere camouflage’ to pursue non-criminal law aims deflect attention from questions concerning the limits of criminalization and leave unchallenged contradictions at the heart of criminal law theory. My purpose in this paper is to examine these offences through some of the basic tenets of criminal law. I argue that the predominant forms of liability for the most often used immigration offences are, at least in principle, controversial and depart from what is often presented as the paradigm in criminal law. Above all, immigration offences are objectionable because they fall short in fulfilling the harm principle and, given that criminal punishment as used against immigration offenders is often a secondary, ancillary sanction to deportation, they license excessive imposition of pain.

1 Introduction

Several countries, such as the United States, Britain, Australia, Germany, France and Italy, have increasingly resorted to criminal law and its institutions to enforce their immigration rules in recent decades. In view of this policy trend, some authors (e.g., Aas 2011; Chacón 2012; Bosworth 2012) have discerned a phenomenon of ‘criminalization of immigration’. While this emerging research field has attracted a great deal of attention by criminologists and socio-legal scholars in the last 15 years or so, in Britain it has progressively become a sort of ‘niche’ for specialized academics genuinely interested in border controls and the regulation of global mobility. By contrast, ‘mainstream’ criminal lawyers have been generally uninterested in the expansive scope of criminal regulation for immigration law-breaking,¹ even though discussions about the limits or boundaries of the criminal law are nowadays in vogue and have generated copious and highly sophisticated outputs (e.g., Duff et al. 2010; Duff and Green 2011; Ashworth et al. 2013).

Elsewhere I have argued that such epistemological division has been unhelpful for developing robust, critical arguments about the unprincipled expansion of immigration-based crimes in recent years and about the use of criminal law in immigration enforcement more generally (Aliverti 2013: 146). Indeed, so-called ‘immigration crimes’, as a category, do not fall into ordinary criminal offences. Nor are they considered as part of the myriad and ever-growing class of ‘regulatory offences’ which have been the focus of much recent academic debate on ‘overcriminalization’ (Stuntz 1996; Green 1997; Ashworth 2000; Husak 2008). The elusiveness of this category of crimes to be accommodated in traditional criminal law taxonomies, compounded by the perception of criminal law scholars that these are not ‘true’ or ‘real’ crimes and hence of no interest to criminal law theory, has resulted in the almost complete lack of scrutiny about the use of criminal law powers in immigration policing.

Unlike other countries in Europe where immigration has been traditionally—and until recently—mostly part of administrative law, Britain has relied for some time on criminal sanctions to deter immigration law-breaking. Early, fragmented rules on ‘alien immigration’ heavily relied on fines, incarceration and expulsion to deal with unwelcome or unruly non-citizens embarking to Britain. The readiness with which contemporary British law-makers have incorporated immigration-based offences in every single piece of immigration and asylum legislation since the late eighteenth century prompts questions for criminal lawyers about the legitimate use of criminal legislation in this field (Aliverti 2012). Surprisingly, though, the deployment of punishment in immigration regulation remained pretty much unquestioned until recently. As Lacey (2009: 952) quite rightly protested, criminal lawyers and philosophers have marginalized ‘peripheral’ offences from their account of formal criminalization because they represent a deviation from paradigmatic or core crimes. Immigration offences are not dealt with in criminal law textbooks; neither are they the subject of more specialized legal writing—not even perfunctorily.

Claims that these offence are not ‘true crimes’ or are a ‘mere camouflage’ to pursue non-criminal law aims deflect attention from questions concerning the limits of criminalization and leave unchallenged contradictions at the heart of criminal law theory. Hence my purpose in this paper is to examine these offences through some of the basic tenets of criminal law: namely, that criminal liability requires an act unless the defendant is under a duty which he fails to fulfil, that he should be at fault before being held liable for wrongdoing (hence the need to establish minimum mens rea requirements) and that the prosecution should bear the burden of proving all the elements of the offence beyond reasonable doubt. In the first part I examine the range of immigration crimes that are currently part of the statute book. My purpose here is not to undertake an exhaustive dogmatic analysis of these offences and the issues they raise. Rather, I want to show that the predominant forms of liability for the most often used immigration offences are, at least in principle, controversial and depart from what is often presented as the paradigm in criminal law. In light of the extensive number of ordinary offences in the statute book which are based on similar controversial forms of liability, in the second part I argue that immigration offences are peculiarly objectionable for two further reasons: they fall short in fulfilling the harm principle and, given that criminal punishment as used against immigration offenders is often a secondary, ancillary sanction to deportation, they license excessive imposition of pain.

2 The Unfairness of Immigration Offences

Given the legal tradition to resort to criminal legislation to regulate foreigners’ movements, it is hardly surprising that the British government, faced with unprecedented levels of asylum applications, resorted to criminal law measures to strengthen border controls from the mid-1900s onwards. During the last Labour government (1997–2010), over 80 new immigration-based offences were enacted. Along with existing offences (for example, illegal entry, overstaying, and facilitation or smuggling), others were created or broadened to criminalize different forms of deception, fraud and non-compliance in the immigration context. There are currently around 95 immigration offences, of which 40 are routinely monitored in Home Office statistics. Most of these offences are barely used: in 2011, 16 of them brought not a single prosecution. Prosecutions (and convictions) cluster round three immigration crimes: seeking leave to enter or remain or postponement of revocation by deception, assisting unlawful immigration, and being unable to produce an immigration document at a leave or asylum interview. Other offences, such as possession of a fraudulent document, are also frequently used against immigration law-breakers.²

Many immigration offences are based on some of the most controversial forms of responsibility in liberal criminal law: strict liability, omission and possession liability. Some of them rely upon ‘situational liability’ and do not require an act. Paradoxically, the leading case on situational liability in English criminal law, R v Larsonneur, deals with the immigration offence of ‘being an alien to whom leave to land in the United Kingdom has been refused was found in the United Kingdom’, in the Aliens Order 1920.³ Not only did this offence lack any mens rea requirement; according to its statutory interpretation, neither did it require a voluntary act by the accused.⁴ In that case the defendant, a French citizen, was escorted from Ireland to the UK by the police. Upon arrival, the British police charged her with the offence under the Aliens Order 1920. Upholding her conviction, the Court of Appeal considered that the circumstances in which the defendant arrived in the UK were ‘perfectly immaterial’ to the charge.⁵ This decision has been harshly criticized by criminal law scholars. Some condemn it because it did not require an act by the defendant. Others argue that there is nothing wrong with imposing situational liability (liability for a state of affairs) as long as the defendant was in a position to control the prohibited situation. Hence they criticized Larsonneur on the ground that it criminalized a state of affairs over which the defendant had no control (Simester et al. 2010: 81; Husak 2010: 45).

Although the offence for which Ms. Larsonneur was convicted has been repealed, many of the existing offences for which non-nationals are often prosecuted are objectionable on a number of grounds. Let us start our analysis with possession offences. Immigration legislation criminalizes a myriad of possession offences: possessing any passport, certificate of entitlement, entry clearance, work permit or other document which the defendant knows or has reasonable cause to believe to be false;⁶ possessing a false or altered registration card without reasonable excuse;⁷ possessing an article designed to be used in making or altering a registration card without reasonable excuse;⁸ and possessing an immigration stamp or a replica immigration stamp without reasonable excuse.⁹ In addition, non-citizens without regular status are prosecuted on a regular basis for possession offences outside immigration laws—simple possession and possession with intent of forged identity documents.¹⁰

Dubber (2001: 935) described this family of offences as a crude manifestation of the ‘police power’, a new version of vagrancy offences and ‘an instrument of nuisance control’. Dubber argued that the creation of these offences has been instrumental to the so-called ‘war on crime’ because they are ‘cleaner, faster, and more convenient’—and, Stuntz (2001: 551) would add, cheaper—to detect and prove. Many possession offences are objectionable because they lack any mens rea requirement and are too remote from harm. Arguably, what are criminalized are not even preparatory acts and therefore are too removed from the actual harm which is sought to be prevented. As Dubber (2001: 864) explained, ‘[m]odern possession liability transfers the danger from the object to its possessor and holds him liable as a source of danger, without the object’s danger ever having manifested itself’. Ashworth (2011) circumscribed these critiques to so-called ‘risk-based’ possession offences, namely those whose rationale for criminalization is the creation of a risk or danger by the possessed object (such as a firearm or weapon). Others (e.g., Tadros 2008; Duff 2009) consider that possession should not be ruled out in principle, attacking instead certain forms of possession liability, namely, simple possession because it criminalizes the mere failure to divest oneself of the relevant object. In particular, Duff (2009: 165) argues that, because simple possession criminalizes someone for what he or she might actually do in the future with the possessed object without requiring proof of intention to do a criminal harm, it infringes the principle of responsible agency.

In the immigration context, charges for possession of forged identity documents result in high conviction rates. Simple possession is relatively easy to detect and prove. The Crown need only prove that the defendant has under his control (i.e., in his luggage or clothes) or is just carrying an identity document which is false, improperly obtained, or that relates to someone else. The possessor of the forged document need not know that the relevant document is false or improperly obtained. Neither does he need to be involved in the forgery or intend to use the relevant document. It is for the possessor to give a ‘reasonable excuse’ on pain of punishment; yet the provision does not specify what sort of explanation might be acceptable. This virtually means that the defendant is under an obligation to provide a ‘reasonable’ explanation, thus shifting the burden of proof onto him. The reversed burden of proof is unfair to the defendant because it requires him to provide an exculpatory explanation to escape punishment even without the prosecution having proved that what he allegedly did is criminally wrong.

The compounded offence of possession with intent requires proof of knowledge or belief that the document is false or improperly obtained, and of ‘improper intention’, defined as either ‘intention of using the document for establishing personal information’ or ‘intention of allowing or inducing another to use it for establishing, ascertaining or verifying personal information about [the defendant] or anyone else’. Possession with intent contains a more demanding standard and carries a substantially higher penalty: a maximum of a 10 year custodial sentence.¹¹ Yet, the prosecution does not need to use the latter, given that a conviction for simple possession is easier and guarantees the automatic deportation of the defendant upon conviction.¹²

Since the possession offences in the Identity Cards Act 2006—now replaced by the 2010 Act—were enacted, prosecutions for offences under the Forgery and Counterfeiting Act 1981 previously used against foreigners with dubious identity documents declined. The offence under section 3 of the 1981 Act requires the use of ‘an instrument’, which the defendant ‘knows or believes to be false, with the intention of inducing somebody to accept it as genuine, and by reason of so accepting it to do or not to do some act to his own or any other person’s prejudice’. It not only requires proof of ‘use’ but also of intention to deceive and of causing actual harm to someone. In turn, the compound possession offence requires possession (under the defendant’s custody or control) of the false instrument plus the intention that he or another ‘shall use it to induce somebody to accept it as genuine, and by reason of so accepting it to do or not to do some act to his own or any other person’s prejudice’. Given patterns in immigration prosecutions, the preference for offences in the 2010 Act over those in the 1981 Act seems to be highly correlated to their ability to secure quick and easy convictions, even if in some cases wrongful ones.¹³

Another general feature of immigration offences is that many of them prohibit not doing. The most often used immigration crime is ‘not having an immigration document at a leave or asylum interview’.¹⁴ Similar offences include: failure to report to a medical officer, failure to comply with conditions of temporary admission, and failure to cooperate with the re-documentation process. Anglo-American legal doctrine considers omission liability as exceptional. This is because to prohibit a particular act (not to do) is less intrusive upon individuals’ autonomy than to mandate a particular act (to do): ‘The burden upon a person enjoined not to do an actus is typically borne more lightly than that when one is ordered to do something, for it involves the sacrifice of fewer options and is more likely to leave the defendant with a chance of conforming without significant derangement’ (Simester 1995: 324). Another reason for restricting liability for omissions is the different moral significance of doing harm as oppose to preventing it, and the importance of preserving such distinction (Duff 2009: 113). ‘Pure’ omissions¹⁵ therefore are adequate basis for criminal liability only when they are attached to a positive duty in virtue of particular roles or relationships: failure to discharge that duty is a crime.

The principle of keeping omissions as an exceptional form of criminal liability is contravened by the offence of ‘no document’. The evil at the core of this offence is the disguise or concealment of foreigners’ identities and the undermining of immigration controls. Instead of being drafted as an act (destroying or disposing of the required document), the actus reus is an omission, which creates a much broader scope for liability while falling short of ‘capturing’ the moral wrong that the offence is supposed to prevent. It is also apt to ask whether the failure to furnish state officials with an immigration document is serious enough to merit public condemnation and punishment.

Another more general objection relates to the ‘source’ of the duty to act whose failure gives rise to criminal responsibility. While citizens’ responsibilities are predicated upon pre-existing legal relations or the very fact of their belonging to a political community, the basis for imposing those duties are less clear in relation to non-member immigration offenders (Duff 1998, 2009: 191, 2011; Zedner 2013). As Laegaard (2010) argued, the state’s right to exclude non-members is precisely predicated upon the fact that uninvited foreigners are not subject of the state’s laws and are prevented from being so. While state-citizen relationships are characterized by ‘mutual recognition and some degree of acceptance of reciprocal rights and duties, the state-immigrant relation is primarily characterized by the relative absence of positive duties on both sides’ (259). It can be argued, however, that in principle non-citizens under the jurisdiction of the host state are both bound and protected by the law of the land, even if their very presence in that jurisdiction is unauthorized, and therefore there is nothing wrong in imposing duties on irregular migrants. In practice, the lack of regular migration status has deleterious effects for non-nationals who find themselves unprotected by the state legal system de facto or de jure—for example, irregular migrants who have been victims of crime may refrain from resorting to the police for fears of being removed (Bucher et al. 2010). In a similar vein, Morales (2015: 1293) argued that, since the state prerogative to inflict punishment relies on the consent by citizens to that sovereign authority, the imposition of criminal liability on unauthorized non-members is illegitimate given that they have no say in the law-making process and no stake in the compliance of those laws. Again, in response to this objection it may be argued that consent theories of punishment are based on the idea of an implicit or tacit consent. Accordingly, by entering and residing in the hosting state, non-members can be said to tacitly consent to the infliction of punishment upon the establishment of their criminal liability.

Although this argument deserves further elaboration than what I provide here, it raises questions on the legitimate basis for imposing and enforcing duties to act which are tightly related to the forbidden access to state’s territory against people whose relationship with that state is (a)legal. While non-nationals should be protected and bound by the law of the host state, and therefore it is legitimate for the latter to expect compliance with its laws, the grounds for imposing positive duties to facilitate their exclusion on unauthorized foreigners is more controversial. Indeed, non-members are under a duty to cooperate in their re-documentation so that their expulsion can take place. Failure to cooperate is an offence. The offence of ‘non-cooperation’¹⁶ imposes criminal liability on those who ‘fail[] without reasonable excuse to comply with a requirement of the Secretary of State’ to provide information which is considered instrumental for obtaining a travel document and thereby to facilitate the defendant’s deportation or removal from the country. This offence does not require knowledge or recklessness as to the failure to comply with state officials’ orders and the courts have been fairly parsimonious in allowing excuses for non-compliance.¹⁷ Forcing people to cooperate with their own removal on pain of punishment is questionable because it distorts the aim of punishment by injecting pragmatic considerations in the use of criminal law. The very object of liability is not to censure a serious wrong, but rather to force the defendant to cooperate. Stipulating negative incentives to obey the law may be thought as one of the legitimate functions of the criminal law. However, such instrumental rationale by itself should not be enough to justify the imposition of criminal liability (Aliverti 2013: 134).

The offence of non-cooperation is also questionable because the imposition of such duty to cooperate is tantamount to self-incrimination in the context of immigration enforcement. Indeed, the privilege against self-incrimination protects the defendant from providing information which can assist the state in proving his guilt, and therefore from the adverse consequence of a criminal conviction and punishment. It can be argued that, by forcing a defendant to provide information instrumental for his removal, the state imposes an unjustified burden on the defendant to assist in his own removal and leaves him unshielded from protection against the adverse consequence of being pushed out, thus infringing a similar protection. While the state may legitimately use its prerogative to expel the non-complying foreigner, it is not entitled to deploy coercive means in order to obtain the cooperation of the deportee to that end. The above discussion does not negate the prerogatives of the state to deny entry or expel those who are not allowed in. Rather, it questions whether such prerogatives can give rise to legal obligations upon the uninvited.

Another pernicious feature of many of these offences is the weak or non-existent mental requirement. The mens rea or fault requirement follows from the respect of the principle of individual autonomy, which in turn demands that in order for the state to legitimately impinge on people’s rights the act or omission which contravenes the law should be free, voluntary and blameworthy. If we just consider the offences in the Immigration Act 1971 (as amended), of 37 offences only eight of them require some fault element (knowledge).¹⁸ Of those with an explicit mens rea requirement, four of them address third parties (people assisting unlawful immigration or helping asylum-seekers, and ship or aircraft captains). The offences most frequently used, such as ‘no document’ and simple possession addressed earlier, do not require mens rea as an essential element. Weak or no mens rea requirement is the rule rather than the exception for these offences.

For the offence of ‘no document’, liability is made out by proving the lack of the required document in the specified circumstances. As a general rule the existence of a fault requirement is presumed when the statute is silent about it.¹⁹ However, the (scarce) case law on this offence has said nothing about it. Not unexpectedly, conviction rates for this offence are extremely high: around 97 per cent of defendants charged with this offence are convicted (Ministry of Justice 2013). Although the specific defence based on the Refugee Convention—akin to ‘duress of circumstances’ or ‘necessity’—is unavailable, people charged with this offence can raise the statutory defences in section 2(4) of the 2004 Act. It is a defence under this section: to prove that the person is an European Economic Area (EEA) national or is a family member of an EEA national exercising Treaty rights in the UK; to produce a false immigration document and prove that he used that document for all purposes in connection with his journey; and to prove that he travelled to the UK without at any stage since setting out on his journey having possession of an immigration document. In addition, to avoid conviction defendants can provide a ‘reasonable excuse for not being in possession of a document of the kind specified’. However, the case law does not provide clear guidance as to what constitutes a reasonable excuse and under what conditions a defendant can escape punishment.²⁰

Further, such burden-shifting onto the defendant—especially since it is a legal burden—is unfair because the prosecution has merely to prove that the defendant does not have an identity document, something that hardly amounts to a criminal wrong. It is unsurprising that this offence produces more guilty pleas than any other immigration offence. While offences of this type (strict liability with reasonable excuse defences) have not been regarded as incompatible with Article 6(2) of the European Convention on Human Rights, the courts have made it clear that in certain circumstances²¹ there can be a breach to the presumption of innocence. In those circumstances, the burden of proof on the defendant should be ‘read down’ to impose an evidential rather than legal burden.²² A legal challenge to the offence of ‘no document’ on this ground was unsuccessful. The Court of Appeal reasoned that because ‘the defendant alone is likely to have all of the relevant information, and bearing in mind the importance of maintaining an effective immigration policy, and the limitation on the penalties which can be imposed under the Act … the burden of proof should be interpreted as being anything less than a legal burden’.²³

The imposition of strict liability is not a feature exclusive to newly created immigration-based offences.²⁴ An examination of their predecessors in repealed statutes shows that in fact they were drafted in an even more draconian manner. Perhaps the harshest provision, which also violated the principle of maximum certainty, was introduced in the Aliens Restriction Act 1914, which penalized ‘any person [who] acts in contravention of, or fails to comply with, any provisions’ of any Order in Council imposing restriction on foreigners.²⁵ Other offences in pre-1971, repealed immigration acts included: to make or cause to be made a false return, statement or representation;²⁶ to refuse to produce or to furnish any information or document required by immigration authorities, or to obstruct them in the exercise of their function; to alter documents or to use or possess for such use ‘forged, altered or irregular certificate, passport, visa or other document’; to enter or remain in contravention of immigration rules; to contravene or fail to comply with conditions or restrictions attached to one’s leave; and to return to the UK in contravention of a deportation order. The Commonwealth Immigrants Act 1968 made it an offence to land in the country without either being examined by immigration authorities upon arrival or submitting to examination, and placed a legal burden on the defendant to produce a passport duly stamped in order to escape liability.²⁷ In contrast, offences penalizing ‘third parties’ usually required some element of fault. So, for instance, the offence of harbouring explicitly required knowledge.

Legal scholars have criticized strict liability offences because they represent a departure from the abovementioned founding principles of liberal criminal law (Ashworth 2009: 161). Yet, some scholars accept strict criminal liability for certain (minor) offences while they regard it problematic for others. Joel Feinberg, for instance, admitted strict liability for certain offences in ‘public welfare statutes’ penalized with fines, while rejecting it for offences punished with imprisonment ‘because imprisonment in modern times has taken on the symbolism of public reprobation’ (Feinberg 1974: 111; Brudner 1993: 31). Similarly, Simester (2005) considered that substantial strict liability should be restricted to non-stigmatic crimes, for instrumental reasons. In contrast he believed that moral objections against holding blameless defendants accountable for ‘stigmatic crimes’ outweigh instrumental arguments in defence for strict liability. While the determination of what counts as ‘stigmatic’ or ‘non-stigmatic’ crimes is unclear, given that immigration crimes are punished with imprisonment and can therefore hardly be classed ‘regulatory’, the general principle that requires proof of fault should hold for these offences. Ashworth (2000: 255) rightly assessed that ‘if a particular wrong is thought serious enough to justify the possibility of a custodial sentence, that wrong should be treated as a crime, with fault required and proper procedural protection for defendants’.

Some of the general features of immigration offences reviewed above—strict and omission liability, and liability for possession—are controversial forms of criminal liability. These features are not ‘exceptions’ to the general rule, but are pervasive and hence distinctive of this family of offences. The most often enforced immigration crimes against foreigners feature these forms of criminal responsibility. Although certain citizens are captured by criminal immigration law, their liability is in a way ‘derivative’ because they are judged for their contribution to the commission of the offence by the principal. Third parties are liable for aiding and abetting an immigration offence. Offences criminalizing third parties (facilitators, smugglers, employers, etc.) are less exposed to the above critiques. Employers who hire foreigners without entitlement to work in the UK are criminally liable only when proof of knowledge of the employee’s status is forthcoming.²⁸ The offences of assisting unlawful immigration and helping an asylum seeker²⁹ also demand proof of knowledge of all the elements of the actus reus. The latter also requires that the helping is done ‘for gain’—and not merely for humanitarian purposes.

These widespread contradictions with principles cherished by contemporary criminal law theory would not be allowed if those rules were to be applied against nationals. Lucia Zedner provides a plausible explanation for the slashing of substantial safeguards in criminal immigration law:

In so far as criminalization rests on the idea that citizens are responsible agents responsive to reasons and that those reasons are ones the individual can fairly be expected to understand by dint of his or her shared membership of law’s community, the very basis for criminal responsibility is attenuated in the case of the non-citizen. Perhaps we should not be surprised, therefore, by the apparent readiness to erode ordinary standards in respect of those to whom no such civic trust is owed and whose very membership of the polity is denied or in doubt. (2013: 52)

A more cynical reading points to the instrumentality of diminished protections in statute-based crimes for ridding society of (unwelcome) foreigners. Indeed, while immigration crimes formally address ‘the abstract foreigner’, in practice the poor, non-white and working class migrant is caught up by these laws (Anderson 2013: 43). This is not only because this group of foreigners is more likely to be involved in immigration crimes than the global elites coming from rich countries (Spena 2013), but also due to more or less explicit policies. In Britain, the Race Relations (Amendment) Act 2000 introduced ‘a limited exemption’ to the prohibition of racial discrimination by public bodies in the exercise of immigration functions, which authorizes differential treatment of nationals of certain countries.³⁰ Although this ‘immigration exemption’ does not cover the investigation and prosecution of immigration offences, the immigration department has recognized that nationals from ‘countries with reasonably high GDP per capita’ tend to breach immigration rules ‘inadvertently’, while others do so deliberately ‘as a way of evading immigration controls’. Hence it concludes that enforcement actions should concentrate on the latter (Home Office 2007: 10).³¹

While the law is formally facially neutral, it is substantially biased as only a particular class of subjects are caught by its premises. Criminal immigration law is ‘a carte blanche for the police controls of undesirables’ (Dubber 2001: 873, 956) and instrumental in ‘the growing war on unauthorized migration’ (Chacón 2012: 614). Norrie (2001: 120) described the criminal law as a ‘formal dance around a set of categories that construct individuals in different ways so as to secure a conviction, and … it is this aim, rather than the prescribed form of the dance, that is determinative’. It is such an aim—the ridding the country of uninvited migrants—that structures the law’s form. Because of their ‘form’, immigration offences and other related crimes are easy to detect and to prove. The most frequently used do not require mens rea and uphold weak actus reus requirements. The defences available are generally narrowly drafted and interpreted restrictively. The economic, political and humanitarian context in which the ‘crime’ is committed is largely left out of sight by circumscribing the judicial examination to the breach of a legal mandate. Those who are proceeded against for these crimes are generally ill-suited to fight back, likely to plead guilty and to be sentenced to a term in prison (Chacón 2009: 140; Eagly 2010: 1321; Aliverti 2013: Ch. 5).

Both the formal and substantial criminalization of these offences is justified on a vaguely defined idea of ‘public interest’, which in turn legitimizes imprisonment and expulsion. In this way, criminal immigration law constitutes a technical means to implement a particular policy objective efficiently. For, if ‘[m]odern liberal law combines in its form individualist right and political necessity’ (Norrie 2009: 25), in this field social control and policy considerations takes precedence over the protection of the individual against state power. And it is precisely the status of that individual which tips the balance against his or her protection.

3 Harm Principle and the Criminalization of Migration Law-Breaking

Immigration offences are based on the most controversial forms of criminal liability and their justification is based on questionable grounds for criminalization. The most frequently prosecuted crimes against non-citizens are based on weak mens rea and actus reus requirements. Indeed, the departure from many criminal law principles is not an exceptional deviation but is an inherent feature of this family of offences. This is not to deny that a large number of criminal offences in the statute book fall short of the criminal law standards reviewed above—particularly driving offences, offences in counter-terrorism legislation and public welfare offences—which reveals that those criminal law principles remain aspirational, rather than descriptive of the actual shape of contemporary criminal law (Ashworth 2009: 137, 2000, 2011: 241).

In the face of the widespread departure from these principles, one may question why immigration-based offences are peculiarly objectionable. I argue that these offences are questionable, first, because they criminalize trivial or harmless wrongdoing and thus fail to fulfil the requirements of the ‘harm principle’. This principle bans criminal proscription in the absence of harm done or threatened. Second, they are objectionable because the intervention of criminal law in this field represents an excessive and unjustified imposition of pain on those subject to it since they are in most cases also liable to expulsion from the country. Indeed, the prospective physical removal of the defendant is in practice intrinsically linked to the type of sanction imposed upon conviction, which is inexorably a custodial sentence (Aliverti 2013: 110; Canton and Hammond 2012: 12).

In terms of the first objection, the justification for these offences generally relies on the negative impact of immigration outside the law. The government argues that immigration outside the law ‘undermines the integrity of the immigration system’ and causes harm to the UK economy, society and individuals, by facilitating welfare abuse, undermining minimum wages and fair competition, creating a ‘pull’ for illegal immigration and leading to more serious crimes, such as terrorism, drug and human trafficking (Home Office 2007: 9, 2010: 12, 18). Criminalization, in this context, seeks to deter people from engaging in immigration wrongdoing which, if left unchecked, can cause harm to the community, public services and the economy. The proposition that unlawful immigration undermines labour competition and overburdens the welfare system, and contributes to organized crime, should be rejected at once, given the poor empirical evidence to support these claims.³² Even if such evidence was forthcoming, criminalization decisions demand more precision on the harm to be prevented in the first place and on the relationship that the prohibited conduct has to that harm or risk of it (Spena 2010: 511).

Ramsay (2012) has argued that the criminal law is increasingly called forth to perform a reassurance function and that criminalization decisions are justified in terms of the protection of citizens’ rights to security. Following this rationale for criminalization, it could be argued that the criminalization of migration breaches pursues that reassurance function by preventing the abuse of migration rules, the overburdening of public services, housing and the labour market, and more importantly the entry of foreigners perceived as threatening and disruptive. In this vein, criminal immigration law is said to perform a harm-preventing function and in doing so it enhances the subjective security of the community of citizens. Ramsay would argue that provided that immigration offences aim at fulfilling that function they satisfy the harm principle—specifically the principle of non-triviality. This is because ‘[o]ntological insecurity is a significant restriction on autonomy … and therefore a significant harm’ (Ramsay 2012: 187). Ramsay’s project is of course to highlight the inadequacy of the harm principle to limit the trend towards overcriminalization observed in contemporary criminal law.

My argument is that we should resist that expansive conception of harm, which runs the risk of turning it in a meaningless principle because the conducts that can be criminalized under it are too far removed from the causation of actual harm. More fundamentally, if we still regard harm as a minimum condition for criminalization, that diluted version of the harm principle can espouse, buttress and legitimize bigoted and prejudiced interests and demands for criminalization that are motivated by genuine or fabricated social anxieties and fears about suspicious others. The embracing of subjective security for guiding criminalization is acutely worrying in the migration control field since it risks licencing the use of criminal law powers territorially excluding undesirable outsiders to placate parochial anxieties, ultimately serving illiberal aims.

The least contested case for criminalization is that immigration law-breaking arguably undermines the efficient running of immigration checks and border controls (Aliverti 2013; Morales 2015: 1280). According to this argument, immigration offences protect a public ‘governmental’ interest. In Feinberg’s taxonomy, these are interests which are ‘generated in the very activity of governing’, such as collecting taxes, registering foreigners, conducting trials and court hearing, etc. The prohibited conducts that these criminal offences aim at protecting harm individuals only indirectly or remotely as they ‘endanger[] the operation of government systems in whose efficient normal functioning we all have a stake’ (Feinberg 1984: 64). Because of this remoteness and in order to preclude the criminalization of trivial interferences or threats to the integrity of the system, Feinberg imposes two limitations to the criminalization of harms to public interests. According to them, criminalization decisions should take into account, first, the ‘extent of the actual or threatened impairment to an institution’s function’ and second, ‘the strength or importance of each individual’s interest in the institution’s health, and the seriousness of the resultant harm when that interest is set back’ (Feinberg 1990: 34).

According to this interpretation of the harm principle, the case for criminalization of the most frequently prosecuted immigration crimes is objectionable. The conducts prohibited by these offences are not closely enough connected to a remote harm to individuals or to the public. Most of them penalize breaches of administrative regulations, and are harmless. They can only have a bearing on the impairment of the system of immigration controls if understood in their ‘accumulative’ or ‘conjunctive’ embodiment: that is, when the conduct criminalized ‘does the feared injury only when combined with similar acts of others’ (von Hirsch 1996: 265). This is an argument commonly resorted to by the courts when sentencing immigration wrongdoers: the conducts penalized (broadly, to gain access to or exit the country through deception) are harmless but have the potential to undermine immigration controls when numerous others follow suit. Their criminalization and punishment is warranted to prevent them from becoming prevalent.³³ In other words, the criminalization of wrongful but harmless conducts is justified because such conducts can become harmful if widely practised, and criminalization will hinder or diminish their occurrence, thus preventing them from becoming harmful.

The argument supporting the criminalization of harmless wrongdoing has been put forward by Gardner and Shute (2000) and Horder (2012), and criticized by many. Horder argued that the criminalization of wrongful conducts that are otherwise harmless is justifiable under the harm principle providing that such conducts can be harmful if they become pervasive and if criminalization is effective in preventing that. Horder (2012: 100) defended the criminalization of harmless conduct when non-criminalization would be harmful because ‘wrongful harm [will become] more common if the conduct is left free from criminal consequences’. One of the main critiques is that this criteria for criminalization fails to pay respect to the autonomy of individual human beings. By placing excessive emphasis on the ends to be achieved—harm reduction, the ‘personal’ wrong is not judged individually but ‘collectively’ together with the wrongs of others and their further ‘cumulative’ effects. As Andrew von Hirsch put it, when the law penalizes a harmless conduct because it becomes injurious when compounded with similar conducts by others ‘the inference from causing harm to doing wrong becomes more tenuous’ (von Hirsch 1996: 265). Tadros (2011: 52) argued that this justification for criminalization is objectionable unless we embrace consequentialism and henceforth accept that the prevention of the greater harm always justifies causing a smaller harm. Finally, there is an empirical objection related to the difficulties—if not impossibility—involved in predicting ex ante a reduction in the incidence of a particular kind of conduct due to criminal proscription. As Ashworth and Zedner (2012: 551) explained, given the shaky evidence on the impact of criminal laws on crime rates, an argument along these lines seems untenable. Hence according to this interpretation of the harm principle, breaches to immigration rules are in principle unsuitable for criminalization, let alone for being punished with imprisonment.

4 Incarceration and Deportation: Excessive and Redundant Punishment

Offences criminalizing unauthorized entry and residence are generally punished with a prison term. In practice, people liable for these offences inexorably receive a custody sentence (Aliverti 2013: 110). The consistent and uniform imposition of custody on convicted migration defendants is not mandated by law. Indeed, section 2(9) of the Asylum and Immigration (Treatment of Claimants, etc.) Act 2004 includes financial sanctions as punishment for the no document offence. Yet, as a matter of penal currency, fines are out of consideration in cases involving migration defaulters.³⁴ As Canton and Hammond (2012: 11) observed, ‘within the legal framework in England and Wales, there are no statutory differences in the powers available to courts when foreign nationals appear before them, but their experiences can be significantly different’, particularly when they are undocumented.

Although there are no clear parameters established in the law to favour custody above other alternatives, section 152(2) of the Criminal Justice Act 2003 clearly stipulates that custodial sentences should not be imposed unless the offence committed is so serious that a fine or a community sentence cannot be justified. The passing of custodial sentences as a matter of course and the almost automatic rejection of non-custodial sentences for migration offences that are non-violent, non-frightening and victimless go against the principle of restraint in the use of imprisonment. Judges generally refer to the seriousness of the offence committed and the need to pass custodial sentences to deter others from undermining the system of border controls. The imposition of a custodial sentence—albeit short in length—for these offences on account of their seriousness contrasts with the consideration of these crimes as ‘low level’ ones by law enforcement agencies and prosecutors, who often regard them as not worth pursuing through criminal charges. The second rationale is questionable on grounds of remoteness (Ashworth 2010: 154). To what extent can we justify the imposition of custody because the defendant’s actions can lead to similar breaches, and thus more serious cumulative consequences, by others? Further, as Ashworth (2010: 241) argued, ‘[t]o pursue preventive strategies through sentencing is as shortsighted as it is unjust. It tends to scapegoat a vulnerable group rather than to seek a long-lasting solution’ to underlying social problems. Deterrence by itself cannot and should not justify disproportionate sentences. As I have argued elsewhere (Aliverti 2012: 426), the justification of punishment—in addition to removal—on deterrence grounds for dealing with immigration defendants is also questionable for another reason: namely, lack of effectiveness. As attested by the daily tragedies in border crossing zones, the threat of criminal punishment is unlikely to have any resonance on people who are so territorially removed from the location where these laws are made and enforced, and for whom the risk of death, interception, incarceration and removal is regarded as a lesser evil in comparison to the dreadful alternative of staying put.

However, other unspoken, pragmatic considerations can explain the judicial preference for custodial sentences in these cases. Non-custodial sentences are generally adjudged to be ill-suited to deal with non-nationals whose crime is precisely to be in the country without authorization and hence are due to be expelled. Indeed, the prosecution, conviction and choice of sanction against immigration wrongdoers are propelled by the ultimate purpose of expelling them from the state’s territory. Immigration law—and the sanctions attached to it—are the primary avenue to deal with unauthorized mobility. Expulsion is repeatedly singled out by policy-makers in Britain and further afield as the crucial measure to impose on those caught attempting to flout immigration rules. Due to their irregular status and their prospective deportation, unauthorized foreign nationals are unlikely to be granted bail and be given non-custodial sentences. For the same reasons, magistrates and judges are unlikely to order a pre-sentence report to inform their sentencing in cases involving foreign nationals, especially if they are undocumented (Canton and Hammond 2012: 9). In other words, the ultimate consequence of entering the country without permission (territorial removal) determines, shapes or, better, taints the form of criminal punishment. Instead of depending on the criminal sanction, as ‘collateral sanctions’ generally do (Ewald 2011), deportation significantly influences, informs and determines the punishment imposed on migration offenders. For them, deportation does not simply follow a criminal conviction. It has a rather crucial function in determining the ultimate punishment.

This inexorable relationship between deportation and punishment flies in the face of basic statutory sentencing principles. The amount and type of punishment ultimately imposed on a convicted defendant should depend on the seriousness of the offence committed, which includes the culpability of the defendant and any harm caused, as established by section 143(1) of the Criminal Justice Act 2003. Further, the use of custody for convicted migration offenders does not pursue rehabilitation goals, one of the aims of punishment generally and of imprisonment in particular. Since foreign national prisoners convicted for migration-related crimes are due to be sent out rather than integrated to the host community, in principle mental health, educational and work training programmes are unavailable for this section of the prison population (Canton and Hammond 2012: 15). Of course, the inexorable connection between criminal punishment and deportation is not unique to migration crimes. Indeed, due to stringent deportation regimes in countries such as Britain and the US (Kanstroom 2000; Bosworth 2011; Chacón 2012), deportation is increasingly an automatic consequence of criminal convictions for foreign nationals. A criminal conviction, as Bosworth (2011: 591) argued, has a ‘more deleterious ramification for non-nationals than ever before, creating a kind of double jeopardy, wherein purely on the basis of citizenship, punishment will effectively vary’. And yet migration offences are particularly vulnerable to critique because they combine the most questionable features of ‘ordinary’ criminal law with the operation of migration-related measures in detriment to those who fall foul of these intertwining regimes.

Whether deportation is regarded as regulatory or punitive, a civil measure or a form of punishment, is beside the point. Its relevance lies not in the legal or bureaucratic classification of deportation but in its material influence for shaping criminal punishment in cases involving immigration defaulters. As I have showed, this material influence in turn results in disproportionate and excessive hardship on people convicted for unauthorized migration.

5 Concluding Remarks

The immigration offences reviewed in this paper are objectionable not only because they are based on the most controversial forms of criminal liability. Their justification under the harm principle is seriously undermined since they criminalize breaches of administrative rules, which are harmless if considered in isolation. Further, the primary sanction against foreigners in breach of immigration laws is expulsion, and the imposition of criminal punishment in addition to expulsion is a disproportionate and unfair consequence of immigration wrongdoing for two reasons: first, because the deportability of the defendant shapes his criminal sentence, thus foreclosing non-custodial options; and second because punishment in this context aims at achieving goals, namely deterrence, which could well be achieved through immigration law sanctions, rendering it redundant and disproportionate.

The regulation of immigration through criminal law is questionable because it imposes disproportionate pain on immigration defendants. If we accept that states have a sovereign right to regulate immigration, and that such right encompasses the prerogative to deny entry to and to eject non-members, then holding unwanted migrants criminally liable for breaching entry and residence rules is simply redundant. Given that immigration breaches are trivial forms of wrongdoing and that expulsion is a measure with drastic consequences for those subject to it, criminalization and punishment—particularly because they often entail imprisonment—inflict disproportionate pain on those subject to it and should be resisted. An application to the principle of parsimony in this context mandates that the state should only use its prerogatives to control migration flows through the regulatory avenue envisaged to achieve that goal—namely immigration laws—instead of multiplying the exercise of state power through different means with the attendant reproduction and exacerbation of hardship on those who fall foul of the dual operation of migration and criminal law powers.