Page 4 of 6
The English Draft Criminal Code
 The Draft Criminal Code proposed by the English Law Commission to a large degree codifies the Tesco decision. 56Corporate liability is directly linked to the commission of the offence by a person who is the company's directing mind. Section 30(2) of the Draft Code provides that "A corporation may be guilty . . . only if one of its controlling officers, acting within the scope of its office and with the fault required, is concerned in the offence." Moreover, the notion of "controlling officer" is limitatively defined, 57 and only persons highly situated in the corporate hierarchy can engage the criminal liability of the corporation. The major criticisms of the overly restrictive nature of the identification theory are not reflected in the English Draft Code. In that sense, the English proposal, were it to be adopted in Canadian law, would constitute a step back from the present situation that would be hard to defend.
The U.S. Model Penal Code
 The Model Penal Code proposed in 1962 by the American Law Institute 58 contemplates three ways in which corporations can be criminally liable. For absolute liability regulatory offences, the principle of vicarious liability is retained. 59 In regard to offences for which the legislature has clearly indicated its intention to adopt corporate liability, the Model Penal Code prescribes a liability regime that is likewise broadly based on the vicarious liability doctrine but includes a possible due diligence defence based on the balance of probabilities, in so far as a "high managerial agent", i.e. someone closely associated with the management of the company, exercised due diligence in an effort to avoid the perpetration of the offence. Finally, in regard to mens rea offences, the model essentially adopts the identification theory as developed in English law. Section 207(1)(c) provides that "A corporation may be convicted of the commission of an offence if . the commission of the offence was authorized, requested, commanded, performed or recklessly tolerated by the board of directors or by high managerial agent acting on behalf of the corporation within the scope of his office or employment."
 The English and American models are variations on the classic themes of vicarious liability and the identification theory. The most innovative aspect of the Model Penal Code lies in the implicit recognition that the actus reus and mens rea of crimes can be attributed to two different persons. To a large degree, however, the problems we identified earlier as to the inappropriateness of these doctrines to serve as an adequate foundation for corporate criminal liability remain unanswered. The traditional doctrines continue to be ill adapted to the context of corporate delinquency, in that they are both too broad and too narrow to be used in sanctioning certain blameworthy conduct, while they allow convictions without proof of genuine fault on the part of the corporation.
The Law Reform Commission of Canada model
 In a working paper presented in 1976, 60 the Law Reform Commission of Canada discussed criminal liability for collective conduct, and came out in favour of corporate liability. Ten years later, the Commission proposed, in its report Recodifying Criminal Law, 61 to more or less codify the applicable law in this regard. The report proposed the following formulation:
2(5) (a) With respect to crimes requiring purpose or recklessness, a corporation is liable for conduct committed on its behalf by its directors, officers or employees acting within the scope of their authority and identifiable as persons with authority over the formulation or implementation of corporate policy.
(b) With respect to crimes requiring negligence a corporation is liable as above, notwithstanding that no director, officer or employee may be held individually liable for the same offence.
 The few comments we made earlier in respect of the English and American drafts can be repeated here. No doubt the most innovative aspect of this draft is its recognition of the collective nature of the wrongdoing, at least of those offences for which the standard of fault is one of negligence. However, as the Commission itself acknowledged, its draft was incomplete:
The problem of diffusion of the elements of a crime among members of the group, discussed above in the context of corporations, also applies to other forms of collective group action. For example, one member of a partnership might do the actus reus, another might have the mens rea, but neither might be liable.. These situations may warrant imposition of criminal liability on the collective. However, this notion of collective responsibility for group action is very complex and we have not been able to formulate any definitive recommendations on this particular issue in our proposed Code. We are of the view that further study on the whole issue of collective responsibility for group action is needed before any radical changes are made in the substance of our criminal law as it relates to this subject.
The second situation not addressed by clause 2(5) nor indeed anywhere in the proposed Code is how far an employer should be liable for the criminal acts of his employee. It is clear that an employer cannot be held responsible for the acts of an employee who goes off on a frolic of his own, unbeknownst to the employer. Much less clear though is the situation where the employer who has control over the employee knows of the employee's criminal activities but stands to benefit from them and acquiesces in them for the purpose of obtaining the benefit. Should there be a positive duty on an employer to prevent such a crime? Or should the employer be liable as a furtherer? This is an issue deserving of further careful consideration. 62
 The English and American drafts, as well as the draft proposed by the Law Reform Commission of Canada, date back several years now, and represent at most an effort to codify some approaches proposed by the case law. In developing the identification and vicarious liability doctrines, the courts have displayed an appropriately pragmatic approach and initiated the discussion on the foundations of corporate liability. However, in Canadian Dredge, Estey J. acknowledged that the vicarious liability and identification doctrines advanced by the courts are not based on any real assessment of corporate personality or of fundamental principles of criminal liability.
This rule [the identification theory] stands in the middle of the range or spectrum. It is but a legal fiction invented for pragmatic reasons.
The position of the corporation in criminal law has been under examination by courts and law makers for centuries. The questions which arise are manifold and complex. They are not likely to be answered in a permanent or universal sense in this appeal, or indeed by the courts acting alone. Proceeding through the history of these issues in the criminal law adds perspective but no clear answer to the problem.63
 Corporate crime is a complex social phenomenon, and in my opinion Parliament cannot be content with codifying the initial responses to it taken by the courts.
The French model
 A consultation of the new French Penal Code 64 yields no solution to the problems that I have been discussing. After extensive discussions on the appropriateness of adopting and codifying criminal liability, the French parliament enacted article 121-2 of the Penal Code, a rather restrictive provision that reads as follows:
[Translation] Bodies corporate, with the exception of the State, are criminally liable in accordance with the distinctions in articles 121-4 to 121-7, and in those cases contemplated by statute or regulation, for the offences committed on their behalf by their instruments or representatives.
 This article, in addition to imposing numerous limitations on corporate criminal liability (since a limited number of corporations could be liable, and only for a prescribed number of offences), effectively codifies the identification theory as developed in the common law jurisdictions. In the first place, corporations are only liable if a particular individual has committed the offence. This individual must be a member of the board of directors or the management of the firm. A mere employee cannot, through his conduct, attract liability in the corporation. 65 The French Code is therefore a long way from developing a notion of collective wrongdoing capable of responding to the criticisms we have been discussing.
 It is therefore necessary, it seems, to turn toward the notion of corporate culture, which has been advanced by some authorities as a foundation for corporate liability.
The notion of corporate culture as a foundation for corporate criminal liability
 It should be noted, first, that the notion of "corporate culture", as advanced by the commentators, has no monolithic meaning. Generally speaking, it can be said that corporate culture refers to a "pattern of shared beliefs and values that give the members of an institution meaning and provide them with the rules for behavior in their organization". 66 This rather broad notion can be used for many purposes, and is helpful in analyzing a corporation's personality in many respects.67 For the purposes of attributing criminal liability, corporate culture refers primarily to the chain of command, the decision-making structure and the general atmosphere concerning obedience to the law. The following indicators are often singled out as pointing to facets of corporate culture that are relevant in the context of criminal liability.
 First, the development within the corporation of clearly defined responsibilities concerning the creation, evaluation and application of standards and procedures designed to ensure compliance with the law by employees would be a significant indicator of a corporate culture that is heedful of compliance with the law. 68 If, for example, the corporate structure is so organized as to deprive senior managers of the information they need to exercise such powers, this would indicate a corporate culture that is designed to elude law enforcement. 69 Generally, deficient structures for the dissemination of information within the firm would also be suspect. 70
 The notion of a corporate culture that encouraged the commission of the offence is also advanced. This notion, especially in the case of very large entities, considers such things as the goals pursued by the firm, the corporate setting, the organizational pressures and the prevailing mentality as factors that may have promoted the commission of an offence.71
 The presence or absence of procedures designed to encourage employees to comply with the law should also be considered.72
 Pamela Bucy, in an important article, discusses in considerable detail all of the indicators that should be considered in determining the corporate culture of a company charged with a criminal offence. Her summary deserves to be quoted at length.
Most of the factfinder's work in applying the corporate ethos standard of liability will occur in the analysis of whether there existed a corporate ethos that encouraged the criminal conduct. The factfinders should examine the corporation's internal structure to make this finding. Beginning with the corporate hierarchy, the fact finders should determine whether the directors' supervision of officers, or management's supervision of employees was dilatory. Next, factfinders should examine the corporate goals, as communicated to the employees, to determine whether these goals could be achieved only by disregarding the law. The third and fourth factors focus on the corporation's affirmative steps to educate and monitor employees and are more relevant in some fields than others. In highly technical fields where corporate employees daily decide issues involving legal compliance or violation, the factfinders should view the corporation's failure to educate its employees as encouraging criminal acts. In other fields where few corporate employees deal in issues affected by law and regulations, the corporation has a minimal duty to educate its employees and its failure to do so is less relevant. In examining the fifth factor, the commission of the present offense, the factfinders should examine the facts considered under the traditional respondeat superior and MPC [Model Penal Code] standards. Unlike these current standards that look to these facts as the sine qua non in imposing liability, however, the corporate ethos standard considers these facts to be relevant, but not conclusive indicia, of corporate liability. The factfinders should assess the sixth factor, how the corporation reacted to past violations, to further evaluate whether the corporation encourages or discourages illegal behavior. Consideration of the last factor, compensation by the corporation, is extremely important because often a corporation's compensation policies most directly influence its employees' behavior. Assessing the message inherent in compensation is complicated because most corporations use at least one form of compensation, indemnification, thus making most corporations criminally liable under the corporate ethos standard. This Article suggests a different approach toward indemnification and insurance coverage of convicted executives: If corporations follow this approach, the factfinder can more fairly weigh this component of a corporation's compensation package. 73
 Finally, it should be noted that the corporate culture or "ethos" can be observed in light of the explicit guidelines issued by officers of the corporation, but that to a large degree it is also and perhaps above all observable from an examination of the institutional practices and implicit policies and rules prevailing within the corporation.
They encompass the routinely tolerated as well as the explicitly sanctioned. This is important given the evidence of tensions between formal rules and informal practices. These tensions are crucial to the understanding of unsafe corporate practice. As French has argued:
"The identity of the central policies of any particular corporation could only be revealed through a careful study of actual corporate behaviour over a period of time. Written statements may be indicative or they may be only window dressing. Acceptance among the corporate personnel or the higher managerial officers determines the content of the policy recognition."
Often there will be no formal corporate licence to break statutory provisions, for example, health and safety regulations. But companies may, by setting off their institutional priorities, create a climate which discourages obedience to known rules. There might be no effective scrutiny of compliance. Here French's stress that it should be the patterns of actual corporate behaviour that are analysed is important. It is not enough to consider a company's positively enacted rules, regulations and instructions. Non-decision making and informal practices must be considered alongside positive acts and institutionalised rules.74
 It must be observed, however, that once the idea is put forward that corporate culture is at the centre of corporate wrongdoing, very few writers propose a corporate fault model that is functional. 75 And even fewer try to put their ideas into some form.
Aggregation of fault elements and fault based on "authority" and "acceptance" (or composite mens rea)
 Some writers cite the Dutch law, which seems to have moved away from the traditional theories of assigning criminal liability to corporations through vicarious liability or the identification theory. See footnote 76 76 The fairly recent cases seem to base corporate liability on two factors: the power of the corporation to determine which acts can be performed by its employees, and the acceptance of these acts in the normal course of business.
The Supreme Court decided that the employee's act could only be regarded as the employer's if:
(1) it was within the defendant's power to determine whether the employee acted in this way, and
(2) the employee's act belonged to a category of acts "accepted" by the firm as being in the course of normal business operations. 77
 A study of the Dutch cases would also suggest that corporate liability could result from the juxtaposition of a set of individual fault elements. 78 This idea of the aggregation of fault elements has sometimes been accepted by the U.S. courts. For example, in United States v. Bank of New England, the Federal Court of Appeal (1st Circuit) held, in regard to knowledge:
(a) collective knowledge instruction is entirely appropriate in the context of corporate criminal liability. Corporations compartmentalize knowledge, subdividing the elements of specific duties and operations into smaller components. The aggregate of those components constitutes the corporation's knowledge of a particular operation.79
 This notion of aggregation of knowledge or individual fault elements as a basis for knowledge or corporate fault is certainly a means of accounting for the complex and sometimes compartmentalized character of the decision-making structures within large firms and the fact that the relevant information is often scattered. However, we must ask ourselves whether this isn't an artificial way of conceptualizing mens rea. It is not hard to conceive of cases in which the guilty intent attributed to a corporation as a result of an aggregation of faults would in fact simply be fictitious. Even in those cases in which there was evidence that the corporation had been negligent in establishing its internal procedures for conveyance of information, it is hard to find recklessness or intention on the basis of such negligence. As Brent Fisse comments:
Composite mens rea is a mechanical concept of mental state that fails to reflect true corporate fault; discrete items of information within organizations do not add up to corporate mens rea unless there is an organizational mens rea in failing to heed them. 80
 Furthermore, the concepts of "power" and "acceptance" advanced by the Dutch cases are vague and imprecise. At most, they are an outline of a model. Field and JÃ¶rg attempted to analyze what exactly was at issue. I will take the liberty of reproducing an excerpt from their study.
It seems likely that acceptance involves judgment on corporate monitoring of risky or illegal behaviour and power is a judgment on corporate response to those risks. It is also clear that there is a normative element to these criteria, that cumulatively they demand an overall judgment on the quality of corporate diligence in establishing, monitoring and enforcing appropriate standards. This is evident in the Hospital case. The management claimed that they could not prevent the unsafe practices because they did not know what was going on. The court's response was that liability was founded on the fact that the management was totally unaware of the routine practices of the hospital and they ought to have been aware of them. Thus "acceptance" does not necessarily involve foresight of the relevant risk. It is not certain whether it extends to any ordinary practice of the business or whether it merely extends to those practices that ought to have been discovered by corporate monitoring mechanisms. The stress above seems to be seen on some notion of reasonableness rather than strict liability.81
 To the degree that these notions refer back to the processes that should have been established in order to avoid the commission of the offence, they appear to be informed by a normative character, and ultimately are used to punish the corporation for its negligence. Clearly, the power and acceptance tests have the advantage of getting away from an analysis of individual conduct and take into account the collective aspect of the fault. However, they appear to allow the attribution to the corporation of offences requiring mens rea, while essentially it is the corporation's negligence that is at issue. In this regard, the notions of power and acceptance, as envisaged, cannot comprehend the corporate recklessness or intent that we were originally attempting to determine. In my view, this is a significant conceptual problem.
"Reactive corporate fault"
 To obviate the latter problem, Brent Fisse has advanced the concept, in recent years, of "reactive corporate fault". 82 Fisse, following the idea put forward by French, recognizes that corporate mens rea is manifested through the corporation's express or implicit policies. However, he recognizes that proof of such policies, especially when they are implicit, is hard to establish, so he suggests that the corporation be given a reasonable opportunity to formulate a policy of compliance with the law after the actus reus of the offence has been committed.
Although strategic mens rea is a genuinely corporate concept of mental state, requiring the prosecution to establish a criminal corporate policy at or before the time that the actus reus of an offence is committed would make corporate mens rea extremely difficult to prove.... The difficulty of proving strategic mens rea, however, may be significantly reduced if the requisite criminal mens rea based on corporate policy need not be shown to have existed at or before the time of the actus reus of the offense. If the corporate defendant is given a reasonable opportunity to formulate a legal compliance policy after the actus reus of an offense is brought to the attention of the policymaking officials, the corporation's fault can be assessed on the basis of its present reactions rather than its previously designed formal policy directives.83
 Fisse argues that gauging the corporation's moral turpitude solely on the basis of attitudes prior to or contemporaneous with the commission of the actus reus obscures the fact that the sometimes inappropriate reactions of companies after they have done something harmful are also blameworthy conduct that is condemned by public opinion. 84 He suggests, therefore, holding companies liable in the event that they fail to undertake remedial measures once the actus reus of an offence has been committed.
Offenses against the person or property, and other specific categories of criminal offenses, could also be converted into offenses of reactive non compliance. This could be done by imposing a general duty on corporations to undertake specified preventive or corrective actions in reaction to having committed the actus reus of an offense, and by making reactive corporate fault a sufficient mens rea. Under this approach, mens rea and actus reus need not be contemporaneous. Inasmuch as the relevant time frame for criminal fault can extend backward to include proactive fault (that is, fault displayed prior to the actus reus), it is difficult to see why the time frame should not also extend forward to include reactive fault.85
 However, in my opinion, the commission of an offence of failing to react correctly after the occurrence of some event, while conceivable, does not settle the question of whether the commission of an initial actus reus in itself constitutes an offence. 86 This can only depend on whether this is some fault prior to or concurrent with the commission of this actus reus. To view the matter otherwise would amount to allowing the corporation a free ride or a "free actus reus". At most, "reactive corporate fault" can serve as proof of intention, recklessness or negligence at the time of occurrence of a second actus reus. But while it is true that the penological objective that is sought primarily through the conviction of corporations is to get them to alter their conduct, it seems to me indefensible, from the standpoint of principle, not to censure the occurrence of an initial actus reus by a corporation when natural persons do not enjoy that advantage. Furthermore, it should be kept in mind that the commission of an initial actus may have devastating consequences that it might prove necessary to punish, without awaiting the occurrence of a second event.
 Incidentally, Fisse's contemplated model necessitates the establishment of structures for the purpose of identifying the anticipated reaction on the part of the organization. 87 In this context, "reactive corporate fault" appears to me to have more to do with contempt of court or a breach of probation, and Fisse's proposed approaches seem more promising when envisaged in a context of expanding the range of possible sentences 88 for corporations or creating a particular offence.
The White Paper
 In June 1993, the federal Minister of Justice tabled a white paper entitled Proposals to amend the Criminal Code (General Principles), which contained some special provisions concerning corporate liability.
 At first sight, these proposals appear to replicate the identification (or delegation) rule as it was formulated by the Supreme Court of Canada. However, a careful reading of the paper indicates some fundamental modifications. By recognizing that a collectivity of individuals may commit an offence, the proposals appear to lean toward recognizing principles of liability peculiar to corporations. But in my view the proposed approach does not extend the logic far enough and is simply a compromise between the adaptation of the traditional liability rules for individuals and the adoption of an original notion of corporate fault.
 Canadian Dredge left unresolved the issue of whether corporate criminal liability should be contingent on a finding of individual guilt. Section 22 of the White Paper answers this question by stating that a corporation may be liable even if the persons who engaged in the criminal conduct or who displayed the appropriate guilty mind are not identified, prosecuted or convicted. This dissociation between individual and corporate liability is even more obvious when one considers that the persons committing the actus reus of the offence and the persons with the requisite guilty mind may be different.
 The identification theory put forward by the Supreme Court in Canadian Dredge required that both the physical and psychological ingredients of the offence be associated with the same individual, at least in regard to mens rea offences. But in the context of large organizations with dispersed operations, those who make the decisions are often isolated from those who execute them. 89 Thus, in recognizing that a collection of individuals may commit the offence, the White Paper takes the first steps toward an original notion of corporate fault. But its proposals are problematic, in my view, in their attempt to preserve the traditional fault spectrum and attribute the guilty mind of an individual to the corporation as the foundation of its liability. The White Paper is still far from grasping the essence of the idea of corporate culture as the foundation of corporate fault.
 In establishing the liability of natural persons, mens rea refers to the cognitive relationship between an individual in regard to the actions he takes, the particular circumstances surrounding his conduct and the consequences that may result. The misconduct is directly linked to the material context in which this person operates. Moreover, the general principles of liability require a close temporal relationship between the physical and psychological ingredients of the crime. The actus reus and the mens rea must, in fact, be concomitant. But once it is recognized that the actus reus and mens rea of an offence may originate with different persons in the corporate context, the requirement of a close psychological and temporal relationship between these two ingredients of the offence necessarily becomes problematic.
 Section 22(1) of the White Paper defines the mens rea of the corporate offence as, in the case of a person having the corporation's express or implied authority to direct, manage or control the corporation's activities in the area concerned, knowing that the act or omission specified in the description of the offence is taking place, has taken place or will take place and having the state of mind required for the commission of the offence. However, the requisite state of mind, as traditionally envisaged, refers to a cognitive process directly linked to the context in which the individual acts. Sections 12.4 and 12.5 of the White Paper, which define the requisite states of mind, refer directly to the act specified in the description of the offence, the circumstances surrounding the perpetrator's conduct and the consequences that may result. Does it follow, then, that the agent having the mens rea must have some knowledge of the exact deed committed by another person and the exact circumstances surrounding that person's conduct? A rigid application of the logic inherent in the principles of individual liability requires an affirmative answer, but at the same time subverts the notion of corporate criminal liability. Furthermore, if the offence can be the deed of more than one individual, it naturally follows that the close temporal relationship that traditionally must exist between the physical and psychological ingredients of the offence should be reconceived. In this sense, the proposed amendments are right in providing for a looser temporal relationship between the ingredients of the offence. However, this abandonment of the rule of concomitance between the actus reus and the mens rea also suggests that the blameworthy state of mind should be redefined other than in close relationship with the particular physical ingredients. To the degree that the mens rea may be prior to the commission of the actus reus by another person, it is obvious that, rather than consisting of a cognitive relationship directly linked to a set of particular facts and circumstances, this mens rea should at most be linked to the commission of an offence understood in its generic sense. In this regard, it seems to us that the wording of the proposals is unclear in so far as it refers to the presence of the requisite mens rea for the commission of the offence.
 Concerning the temporal relationship between the actus reus and the mens rea, I also wish to draw attention to the fact that the proposal to adopt corporate liability by associating a state of mind to a previous event is particularly problematic. It is questionable, for example, what the intention, as defined in section 12.4, might correspond to in relation to an event that has already occurred. It is hard to imagine how one can wish the occurrence of an event that has already happened other than by its acceptance a posteriori. To base guilt on some misdeed subsequent to the commission of the actus reus risks, in my view, associating intention with the passive acceptance of a result or a mere failure to take remedial steps. The boundary between negligence, recklessness and intention, if not impossible to define in such cases, is certainly a hard one to establish on the facts.
 This option of stretching the temporal relationship to incorporate in the offence a mens rea subsequent to the commission of the actus reus is not unrelated to the notion of "reactive corporate fault" on which Professor Brent Fisse has been working for a number of years, as we noted earlier. 90 The White Paper proposals constitute an initial step toward the recognition of a notion of corporate fault, but there are some problems in the attachment to individual cognitive processes as the foundation for corporate liability. The proposed model could be completed by drawing on the Australian Criminal Code. The provisions of that Code constitute, in my opinion, the most accomplished formalization of a notion of true corporate fault.
The Australian model
 The Parliament of Australia adopted a new Criminal Code in 1995. Division 12, which is addressed to corporate criminal liability, constitutes an original effort to adapt the general principles of criminal liability to the particularly complex circumstances of corporations. The drafters attempted to develop a notion of corporate fault that reflects the diffuse nature of the decision-making process in large companies, drawing extensively on the recent work of authorities such as Fisse who are trying to develop a fault model based, inter alia, on the observed functioning of corporate entities 91 and the notion of corporate culture as the foundation for their liability. The notion of "corporate intention" is not reducible to the individual intention of employees, managers or officers. Rather, it corresponds to the express or implicit policies governing the company's activities. The Australian Criminal Code attempts to integrate these notions. The result deserves to be quoted in full and is reproduced in an appendix to this study.
 The Division 12 provisions inspire a number of comments. First, the structure of the provisions is interesting, in that it deals separately with the actus reus and the mens rea. The various ingredients of the offence may be associated with more than one individual. In this regard, the approach resembles that of the White Paper. Particularly attractive, in my view, is the treatment of the actus reus in a special provision applicable to all offences, with another section devoted to the various fault elements. The separation between the physical elements and the mens rea is apparent just from reading the provisions, without the need to expressly refer to it. Such an arrangement would offer the advantage, in Canadian law, of including in the Criminal Code a provision applicable to all offences, whether criminal or regulatory. In my opinion, a codification of the corporate criminal provisions should be sufficiently general to be applicable to all types of offences.
 It should be noted, however, that the Australian Criminal Code, by attributing to the corporation all of the acts committed by its employees or agents acting within their actual or apparent scope of authority, does not incorporate any real notion of corporate acts. In so far as the commission of the actus reus of offences is concerned, the corporation's liability is based essentially on the theory of vicarious liability. At most, the limitation of the mistake of fact defence in section 12.5 constitutes an attempt to distinguish the acts of individuals from acts that are strictly corporate. However, the questions we raised earlier concerning the availability to corporations of the defences related to the voluntary aspect of the actus reus that are available to individuals remain largely unanswered.
 Concerning the mens rea of offences, it will be noted, first, that the range of faults is retained. A corporation may be charged with an offence involving intention, knowledge, recklessness or negligence. In this regard, the Australian Code attempts a clear distinction between subjective fault elements and negligence.
 The procedure it establishes is, prima facie, rather complex. Section 12.3 provides that if intention, knowledge or recklessness is an essential ingredient of the offence, these fault elements must be attributed to the body corporate if it expressly, tacitly or impliedly authorized or permitted the commission of the offence. The section then prescribes four means by which such an authorization or permission may be established. 92 First, the corporation's fault will be established if the body corporate's board of directors intentionally, knowingly or recklessly carried out the wrongful conduct, or expressly or by necessary implication authorized or permitted the commission of the offence. Second, the corporation's fault may be established by evidence that a high managerial agent of the company intentionally, knowingly or recklessly engaged in the relevant conduct or expressly, tacitly or impliedly authorized or permitted the commission of the offence. In this second case, however, the corporation will not be liable if it proves that it exercised due diligence to prevent the conduct. Third, the corporation's fault may be established by proof that a corporate culture existed within the body corporate that encouraged, tolerated or led to non- compliance with the relevant provision. Fourth, the corporation's fault may be established by proving that it failed to create and maintain a corporate culture that required compliance with the relevant provision. We note, finally, that section 12.3 specifies that if recklessness is not a fault element in relation to a physical element of an offence, the corporation's liability cannot be established by proof of recklessness by the board of directors or a high managerial agent of the company.
 "Corporate culture" is defined in subsection 12.3(6) as "an attitude, policy, rule, course of conduct or practice existing within the body corporate generally or in the part of the body corporate in which the relevant activities takes [sic] place".
 To avoid the automatic attribution to the corporation of law-breaking through personal initiatives by its officials, subsection 12.3(3) provides that the corporation may escape conviction if it proves that it exercised due diligence to prevent its officers from intentionally committing or permitting the offence.
 Corporate negligence is established by proof of negligence of its employees, agents or officers or, if no one individually is negligent, that the body corporate's conduct, viewed as a whole, is negligent. This collective negligence may be established by proof that the prohibited conduct was substantially attributable to inadequate management, control or supervision, or failure to provide adequate systems for conveying information within the body corporate.
 Thus the Australian Code takes a big step toward the definition of a notion of corporate fault. In this regard, the notion of corporate culture that encouraged the offence is of particular interest. This notion, especially in the case of extremely large entities, accounts for things in the corporate setting, organizational pressures and prevailing mindset that may have encouraged the commission of the offence. The notion of corporate culture can be used to convict a corporation even though no clear mens rea can be identified in a particular individual, and the notion serves as a particularly clear expression of the collective aspect of corporate fault. The concept of corporate culture is an original response to the frequent criticism that the identification theory is too restrictive to actually account for corporate fault. Moreover, giving the corporation the possible defence that it exercised due diligence to prevent the commission of the offence is a means of precluding the conviction of the company in the absence of any real fault on its part by automatically attributing to it the fault of an individual. This mitigation of the identification theory thus provides an answer to those who criticize that theory for casting too wide a net. The courts in Canada have clearly attempted to steer a middle course between overly broad and overly narrow interpretations of corporate liability through a tedious exercise of determining which individuals, under the identification theory, can engage the corporation's liability. The Australian Code seeks to achieve a balance in another way. While it still leans on the identification theory since a corporation is prima facie liable for the commission of an offence by a high managerial agent it expands the notion of fault through the notion of corporate culture, while mitigating it through the defence of due diligence. Corporate fault is therefore broadly envisaged as a collective notion.
 The Australian Code clearly constitutes the most serious and refined effort to formalize a notion of genuine corporate fault. However, it highlights a fundamental conceptual difficulty inherent in the notion of corporate culture. Whatever the efforts made to maintain a clear distinction between subjectively assessed faults and negligence, it seems hard to avoid consistently coming back to negligence as the true foundation for corporate liability. Indeed, providing that a deficient corporate culture can be the basis for a charge of intentionally committing a crime transforms into an intention what in my opinion is simply negligence. In my opinion, while the corporate culture can be the basis for a corporation's criminal liability, a corporation surely cannot be convicted of a crime of intention simply by proving that a deficient corporate culture led to the commission of the offence, or that the company was deficient in failing to maintain a corporate culture that encouraged respect for the law. It must be proved that the corporate culture instigated, encouraged or led to the commission of the offence or that the failure to maintain a law-abiding atmosphere was deliberate.