The Humane League UK, R (on the application of) v Secretary of State for Environment, Food and Rural Affairs

This is an appeal from the decision of Sir Ross Cranston dismissing the appellant’s claim for judicial review in which it sought a declaration that the Secretary of State’s policy permitting the keeping of fast-growing breeds of chicken and his system for monitoring their welfare are unlawful. That claim was founded on a submission that the Secretary of State had misinterpreted paragraph 29 of Schedule 1 to the Welfare of Farmed Animals (England) Regulations 2007 (‘Paragraph 29’), which provides:

‘Animals may only be kept for farming purposes if it can reasonably be expected, on the basis of their genotype or phenotype, that they can be kept without any detrimental effect on their health or welfare.’ (Footnote: ¹)

On this appeal Mr Edward Brown KC for the appellant now accepts that this court cannot determine whether the keeping of such fast-growing breeds is unlawful. That depends on contested scientific evidence and is not appropriate for decision on a claim for judicial review – particularly in the absence of any challenge to the lawfulness of the Secretary of State’s policy on the ground that it is irrational. It follows from this that the claim as pleaded must fail. However, Mr Brown invites us to rule on the scope and effect of Paragraph 29 so that the Secretary of State may consider whether his policies and the Code of Practice issued by him in 2018 require revision.

I am reluctant to give such a ruling in the abstract. However, the judge has expressed some conclusions about the meaning of Paragraph 29 and we have heard full argument about it. It is therefore necessary, in my view, for us to say something about this, and it will be convenient to test the rival interpretations by reference to some of the evidence about fast-growing chickens which has been deployed before us. But it needs to be understood by any reader of this judgment, not least because of the strong feelings which this subject excites, that we are not in a position to reach any conclusions about that evidence. That will be a matter for the Secretary of State to consider in accordance with usual public law principles – or, if necessary, for a magistrates’ court to which Parliament has entrusted any prosecution for breach of Paragraph 29.

The appellant is a charity whose mission is to end the abuse of animals raised for food. It is supported in these proceedings by the RSPCA, described by the judge as the world’s oldest and largest animal welfare charity.

The respondent is the Secretary of State for Environment, Food & Rural Affairs. Among other things he exercises powers contained in the Animal Welfare Act 2006, makes regulations under that Act, and issues Codes of Practice concerning the welfare of farmed animals.

The Secretary of State’s evidence is that more than one billion meat chickens (or ‘broilers’) are slaughtered in the UK every year and that in 2021 the overall value of meat chicken production in the UK was £2.4 billion. The overwhelming majority of these chickens are produced in what has been called an ‘integrated model’, whereby the farmer typically owns the land and buildings and is contracted to produce chickens for an ‘integrator’. The integrator manages the slaughter of the chickens, their processing and packing, and contracts with retailers such as supermarkets. The market is highly concentrated, with the three main integrators supplying approximately 75% of all meat chickens.

Approximately 95% of meat chickens are reared in large, closed buildings with a maximum stocking density of 39 kg per square metre, although some retailers may stipulate for lower stocking densities. Chickens reared in this way are described in the legislation as ‘conventionally reared’.

These proceedings concern what are called ‘fast-growing chickens’. These are chickens which, as a result of genetic selection, can be expected to reach a slaughter weight of about 2.2 kg in 5-6 weeks when conventionally reared (and all fast-growing chickens are conventionally reared). There are several different breeds of such fast-growing chickens, the three main breeds being the Cobb 500, the Hubbard Flex and the Ross 308. By way of comparison, slow-growing breeds, which have not been genetically selected in this way, can be expected to reach their slaughter weight in 7-8 weeks when conventionally reared. However, only a small percentage of conventionally reared meat chickens in the UK are slow-growing breeds.

In 2020 the RSPCA produced a report on welfare standards for meat chickens entitled ‘Eat. Sit. Suffer. Repeat. The Life of a Typical Meat Chicken’ (‘the RSPCA Report’). It was this report which led the appellant to launch its claim for judicial review, although the issues relating to fast-growing chickens are of long standing (see e.g. R v Compassion in World Farming Ltd v Secretary of State for the Environment, Food and Rural Affairs [2004] EWCA Civ 1009). The contents of the RSPCA Report were summarised by the judge in the following terms:

‘18. The executive summary of the report states that the genetic selection of meat chickens for performance:

“has been reported to be responsible for contributing to not only the most, but also the most severe, welfare problems seen in broilers today, such as chronic leg disorders and heart and circulatory problems. The severity of the welfare problems, the huge number of animals involved globally, and the fact that these welfare concerns have not been adequately addressed to date, means this long-standing issue requires urgent attention.”

19.

After an overview of the broiler genetics industry, the report has a section on the welfare implications of intense genetic selection for performance. The first heading is “health”, and the report has three subsections. The first addresses heart and circulatory health and states that fast growth can increase the risk of two types of heart conditions, ascites, and sudden death syndrome. There is reference to several studies to support and explain this. Walking ability is the next subsection, and how fast growth can cause leg developmental disorders. Again the matter is developed by reference to relevant research on leg fractures and walking ability. The final sub-section, hock burn and foot burn, states that prolonged periods of inactivity can contribute to the development of ulcers and lesions on those areas of the bird that are in contact with the floor, typically the feet (foot burn) and hocks (hock burn).

20.

Following the section on health the report has a section on behaviour. Selective breeding for increased performance, it states, has resulted in a reduction in the activities the birds can carry out. It then explains that healthy chickens are motivated to perform a wide range of behaviours, including foraging, dustbathing, and perching, all of which contribute to good welfare. If a chicken cannot express a full repertoire of natural behaviour, the report continues, it may experience frustration, helplessness or boredom and may not have the opportunity to experience pleasure or other positive states. There are then sub-sections on foraging, dust bathing and perching.

21.

For the purposes of the report, the RSPCA commissioned research by Dr Laura Dixon at Scotland’s Rural College to assess the production and welfare characteristics of fast-growing meat chickens (conventional breeds) compared with a slow growing breed: “Slow and steady wins the race: The behaviour and welfare of commercial faster growing broiler breeds compared to a commercial slower growing breed” PLoS One 2020;15(4):e0231006.

22.

The RSPCA report summarises the findings of Dr Dixon’s research as follows:

“The trial revealed that, in general, compared to the slower growing breed, the conventional breeds had significantly poorer health: higher mortality (including culls), poorer leg, hock and plumage health, and more birds affected by breast muscle disease (white striping and wooden breast) ... The conventional breeds were also less active, spending less time walking and standing, and more time feeding and sitting, and spent less time engaged in enrichment type behaviours: foraging, perching and dust bathing”.’

It should be said that these conclusions are controversial and that the validity of the trial on which they are based is challenged. We cannot resolve these issues and have not been asked to do so. In addition the Secretary of State makes the point that all breeds of conventionally reared chickens can be expected to experience some degree of heart and leg problems.

Paragraph 29 is contained in Regulations made under section 12 of the Animal Welfare Act 2006, but it has its origin in European Union law, specifically Directive 98/58/EC, the ‘Farming Directive’. The Farming Directive applied to all animals bred or kept for farming purposes and laid down common minimum standards for their protection. Article 4 provided that Members States should:

‘ensure that the conditions under which animals … are bred or kept, having regard to their species and to their degree of development, adaptation and domestication, and to their physiological and ethological needs in accordance with established experience and scientific knowledge, comply with the provisions set out in the Annex.’

Under the heading of ‘Breeding procedures’, paragraphs 20 and 21 of the Annex provided:

‘20. Natural or artificial breeding or breeding procedures which cause or are likely to cause suffering or injury to any of the animals concerned must not be practised.

This provision shall not preclude the use of certain procedures likely to cause minimal or momentary suffering or injury, or which might necessitate interventions which would not cause lasting injury, where these are allowed by national provisions.

21.

No animal shall be kept for farming purposes unless it can reasonably be expected, on the basis of its genotype or phenotype, that it can be kept without detrimental effect on its health or welfare.’

Directive 2007/43/EC, the ‘Chicken Directive’, dealt specifically with chickens kept for meat production by laying down minimum rules for their protection. It dealt with such matters as feeding, litter, ventilation and heating, noise, light, inspection, cleaning, record keeping and stocking density.

Turning now to the national law with which we are concerned, section 4(1) of the Animal Welfare Act 2006 (‘the 2006 Act’) makes it an offence for a person who is responsible for an animal to cause that animal unnecessary suffering, while section 9(1) provides that a person commits an offence if they do not take reasonable steps in all the circumstances to ensure that the needs of an animal for which they are responsible are met to the extent required by good practice. Subsection (2) goes on to provide that for the purposes of the Act, an animal’s needs shall be taken to include (a) its need for a suitable environment, (b) its need for a suitable diet, (c) its need to be able to exhibit normal behaviour patterns, (d) any need it has to be housed with, or apart from, other animals, and (e) its need to be protected from pain, suffering, injury and disease.

Section 12 of the 2006 Act provides that the appropriate national authority (in England, the Secretary of State) may make regulations for the purpose of promoting the welfare of animals for which a person is responsible. The power to make such regulations includes power to provide that breach of a provision of the regulations is an offence.

Section 14 of the 2006 Act confers on the Secretary of State the power to issue codes of practice for the purpose of providing practical guidance in respect of its provisions:

‘(1) The appropriate national authority may issue, and may from time to time revise, codes of practice for the purpose of providing practical guidance in respect of any provision made by or under this Act. …

(3)

A person's failure to comply with a provision of a code of practice issued under this section shall not of itself render him liable to proceedings of any kind.

(4)

In any proceedings against a person for an offence under the Act or an offence under regulations under section 12 or 13—

(a)

failure to comply with a relevant provision of a code of practice issued under this section may be relied upon as tending to establish liability, and

(b)

compliance with the relevant provision of such a code of practice may be relied upon as tending to negative liability.’

Sections 15 and 17 of the 2006 Act set out the process to be followed when the Secretary of State proposes to issue, revise or revoke a code of practice. They include provision for consultation with interested persons.

Mr Richard Turney KC for the Secretary of State insisted that the Secretary of State is under no duty to issue any code of practice or to give any guidance concerning regulations issued under section 12. That may be so, although the legislation appears to contemplate that the Secretary of State will issue codes of practice and he has in fact issued such codes relating to all of the main kinds of animals kept for farming purposes, including meat chickens.

The Welfare of Farmed Animals (England) Regulations 2007, SI 2007/2078 (‘the 2007 Regulations’) were made under section 12 of the 2006 Act. They implemented Article 4 of the Farming Directive by imposing a duty on a person responsible for a farmed animal to take ‘all reasonable steps to ensure that the conditions under which the animal is bred or kept comply with Schedule 1’: regulation 4(1). In complying with this duty, a person must have regard to the animal’s (a) species, (b) degree of development, (c) adaptation and domestication, and (d) physiological and ethological needs ‘in accordance with good practice and scientific knowledge’.

Regulation 5(1) imposes additional duties on persons responsible for certain kinds of farmed animal. A person responsible for conventionally reared meat chickens must comply with Part 2 of Schedule 5A. The provisions in Part 2 of Schedule 5A correspond to those in the Chicken Directive.

Regulation 6(1) requires that a person responsible for a farmed animal must not attend to the animal unless he is ‘acquainted with any relevant code of practice …’ and must take all reasonable steps to ensure that a person employed or engaged by him does not attend to the animal unless acquainted with any such code.

Regulation 7(1) creates an offence if, without lawful authority or excuse, a person contravenes a duty in regulation 4, 5 or 6, or causes or permits this to occur. Regulation 8 provides that a local authority may prosecute proceedings for an offence under the Regulations, but also enables the Secretary of State to direct that he (or the Director of Public Prosecutions) and not the local authority will prosecute ‘in relation to cases of a particular description or any particular case’.

Schedule 1 to the 2007 Regulations, which implements the Annex to the Farming Directive, contains general conditions under which farmed animals must be kept. It deals with staffing, inspection, lighting, drainage, sickness and injury, record keeping, freedom of movement, buildings and accommodation, animals not kept in buildings, automatic or mechanical equipment, and feed, water and other substances.

Paragraphs 28 and 29 appear under the heading ‘Breeding procedures’. They correspond to paragraphs 20 and 21 of the Annex to the Farming Directive. They provide:

‘28. (1) Natural or artificial breeding or breeding procedures which cause, or are likely to cause, suffering or injury to any of the animals concerned must not be practised.

(2)

Sub-paragraph (1) does not preclude the use of natural or artificial breeding procedures that are likely to cause minimal or momentary suffering or injury or that might necessitate interventions which would not cause lasting injury.

29.

Animals may only be kept for farming purposes if it can reasonably be expected, on the basis of their genotype or phenotype, that they can be kept without any detrimental effect on their health or welfare.’

I shall refer to the words which I have italicised as ‘the no detriment proviso’.

While Schedule 1 applies generally to all farmed animals, Schedule 5A deals specifically with conventionally reared meat chickens. It was added to the 2007 Regulations in 2010 by the Welfare of Farmed Animals (England) (Amendment) Regulations 2010, SI 2010/3033, to implement the Chicken Directive. It contains provisions dealing (among other things) with maximum stocking density, access to food and water, litter, ventilation and heating, noise, light, inspection, cleaning and record keeping. It is concerned, therefore, with the conditions in which conventionally reared meat chickens may be kept and not with the breeds which it is permissible to keep.

Part 3 of Schedule 5A provides for monitoring of the chickens’ mortality rate on arrival at the slaughterhouse in order to identify possible indications that the chickens have been kept in ‘poor animal welfare conditions’. Paragraph 15 provides:

‘(1) An official veterinarian conducting controls under Regulation (EU) 2017/625 in relation to chickens must evaluate the results of the post-mortem inspection to identify possible indications of poor welfare conditions in their holding or house of origin.

(2)

If the mortality rate of the chickens or the results of the post-mortem inspection are consistent with poor animal welfare conditions, the official veterinarian must communicate the data to the keeper of those chickens and to the Secretary of State without delay.’

The Secretary of State has issued a Code of Practice for the welfare of meat chickens and meat breeding chickens pursuant to section 14 of the 2006 Act. The current version was issued in 2018. The preface to the Code of Practice states that it is intended to help those who care for meat chickens and meat breeding chickens to practise good standards of stockmanship to safeguard chicken welfare. The Code sets out various provisions of the legislation in coloured boxes and contains numerous recommendations.

Paragraph 29 is referred to in two places. First, in the introduction, the terms of Paragraph 29 are set out in a coloured box, followed by a recommendation that ‘the strains of bird selected must be suitable for the production system’. Second, Paragraph 29 is set out again in a section of the Code of Practice dealing with additional recommendations for ‘meat breeding and grandparent chickens’. Such chickens are not kept for meat production and this reference can therefore be ignored for the purpose of this appeal. Finally, although Paragraph 29 of the 2007 Regulations is not referred to at this point, paragraphs 58 and 59 of the Code of Practice, dealing with the leg health of meat chickens, state that:

‘58. Leg disorders with associated lameness can be a key cause of poor welfare in meat chickens. There are many causes of leg disorders leading to poor leg health including those linked to nutrition, microbial infection and genetics. Nutritional deficiencies and imbalances including calcium, phosphorus and vitamin D can lead to an increase in bone deformities and lameness. Lameness can also be caused by bone or joint infection, so effective prevention and control of viral and bacterial disease plus good litter management are essential.

59.

Welfare and health considerations in addition to productivity, should be taken into account when choosing a strain for a particular purpose or production system. In line with this, meat chickens should stem from broad breeding programs, which promote and protect health, welfare and productivity. Keeping birds in line with appropriate growth curves that optimise these criteria, particularly with regard to leg health, should be considered.’

Pursuant to the monitoring requirements in Part 3 of Schedule 5A to the 2007 Regulations, the Secretary of State has developed what is called the ‘trigger system’. It is intended to identify poor welfare conditions in conventionally reared meat chicken holdings (the production site where chickens are kept) or houses of origin (the building on a holding where a flock of chickens is kept).

The trigger system was designed in consultation with animal welfare organisations including the RSPCA and has been reviewed over time. As described in the Code of Practice, all batches of conventionally reared meat chickens are monitored at the slaughterhouse and the results are used by the Animal and Plant Health Agency to identify possible welfare problems at the farm. The system operates by assessing the results of ante-mortem and post-mortem inspections carried out at slaughterhouses to see whether levels of a range of conditions or mortality rates are unusually high. If predetermined thresholds, known as ‘trigger levels’, are exceeded, a report is generated and sent to the keeper of the birds. The reports are then used to identify farms at the highest risk of non-compliance, so that farm inspections can be appropriately targeted.

The judge recorded that, according to the evidence, less than 5% of meat chicken batches generate a trigger report each year, and that reports do not necessarily generate an investigation or a review of conditions on the farm.

The judge noted the Secretary of State’s position that the appellant’s case had changed during the course of the proceedings and that the appellant was out of time for any challenge in relation to the Code of Practice or the trigger system. However, he decided not to deal with these procedural and timing objections, but to address the appellant’s substantive challenge. There is no respondent’s notice from the Secretary of State and Mr Turney did not suggest that we should do otherwise. In a way that is unfortunate. A claim for judicial review ought as a minimum to identify the decision which is challenged so that the argument can be appropriately focused. This modest requirement of procedural rigour promotes clarity of decision-making, not only by a public authority but also by the court, while the time limit for bringing a judicial review claim promotes orderly administration. As it is, we are left with a dispute about the meaning and effect of Paragraph 29 which is not focused on the lawfulness of any particular decision, in circumstances where the judge addressed its meaning in his judgment but did not crystallise this in any formal declaration, understandably taking the view that in the absence of a clear factual matrix it was not appropriate for the court to make a declaration in the abstract.

The judge rejected the appellant’s case that Paragraph 29 should be read as containing a prohibition on the keeping of farmed animals subject only to what I have called the no detriment proviso. He considered that this interpretation was ruled out because a breach of Paragraph 29 is a criminal offence, so that the burden would be on the keeper of any farm animal to satisfy a criminal court as to the no detriment proviso, and that a narrower interpretation of the paragraph was therefore required. As to that proviso, the judge held that the standard of reasonableness is objective, and that the person whose expectation is relevant is a reasonable person responsible for the animals, who need not have knowledge of scientific literature or such matters as the RSPCA Report. He held that the words ‘any detrimental effect’ connoted ‘obvious or deleterious harm’, so that in summary:

‘94. Consequently, my view is that Paragraph 29 means what it says given the matters referred to above. Given the breed of animal chosen for keeping for farming purposes, it must reasonably be expected by the reasonable person responsible for them that given their genotype or phenotype they can be kept in appropriate conditions without any obvious or deleterious effect on their health or welfare.’

The judge then rejected the appellant’s challenge to the Code of Practice. The Secretary of State was under no obligation to inform keepers that they were prohibited from keeping fast-growing chickens and the Code of Practice did not indicate that the Secretary of State had a policy that fast-growing breeds of meat chicken could be kept. It did not provide any specific advice as to the breeds of meat chicken which could be kept and did not offer any interpretation of Paragraph 29. The Secretary of State had taken expert advice and accepted that there may be a higher risk of welfare issues with fast-growing meat chickens, but took the view that they could be kept without detriment to their welfare since environmental conditions could have an influence on the health and welfare of birds with both fast- and slow-growing breeds.

Finally, the judge rejected the appellant’s challenge to the trigger system. None of the appellant’s criticisms arose out of any error in the interpretation of Paragraph 29 and they did not meet the high threshold for a successful irrationality challenge, especially in a context where technical experts could differ.

It is useful to focus on the appellant’s grounds of appeal, which were as follows:

Despite the complaint that the judge ought to have issued a declaration, it was only in the course of the hearing before us, and even then only at the request of the court, that the appellant formulated the declaration which it now invites us to make. This turned out to be in the following terms, which encapsulate Mr Brown’s submissions:

‘The Court declares that:

(1)

Paragraph 29 concerns the choice of breed (by reference to genotype or phenotype) which may be kept.

(2)

“Kept for farming purposes” means that the animal is of a kind ordinarily kept as a farmed animal.

(3)

“May only” means that the keeping of animals is prohibited unless the conditions of paragraph 29 are satisfied.

(4)

“Reasonably be expected” means what would be expected by a keeper operating in ordinary commercial farming conditions.

(5)

The “genotype or phenotype” (i.e. breed) of the animal must be the cause of the detrimental effect.

(6)

“Detrimental effect” means any kind of detriment that has an effect on the health and welfare of the animal which is more than de minimis.

(7)

If there is “any” relevant detriment then the breed cannot be kept.’

The Secretary of State’s position is that, reading the language of Paragraph 29 in its statutory context, it is a prohibition on the keeping of farmed animals whose genotype and phenotype mean that, regardless of the conditions in which they are kept, they cannot be kept without detriment to their health or welfare, a matter which necessarily requires an evaluative judgment informed by veterinary science; and that there is no scientific consensus or conclusive evidence that fast-growing meat chickens cannot be kept without such detriment. The Secretary of State takes the view that they can be kept without detriment to their welfare since environmental conditions can have an influence on the health and welfare of birds with both fast- and slow-growing breeds. As I understand it, the Secretary of State’s position is, therefore, that any detriment to health or welfare suffered by fast-growing chickens as a result of the way in which they have been bred can be mitigated by the environmental conditions in which they are kept.

It is common ground that the meaning of Paragraph 29 is at the heart of this appeal.

As the meaning of Paragraph 29 depends on a process of statutory interpretation, the search is for the meaning of the words used in the particular context of the legislative scheme which I have described, including in particular the 2007 Regulations, taking account of the paragraph’s origin in European Union law. That context includes, in the case of meat chickens, an acknowledgement that the overwhelming majority of such chickens are reared and spend their whole lives in large, closed buildings. This acknowledgement is implicit in the use of the phrase ‘conventionally reared meat chickens’. It is implicit also that the legislation does not regard such conditions, provided that there is compliance with the minimum standards set out in Schedule 5A, as inherently detrimental to a meat chicken’s health or welfare.

The context also includes an acknowledgement that inevitably some birds will die or experience bad health before they arrive at the slaughterhouse. That is implicit in Part 3 of Schedule 5A which takes it for granted that there will be a ‘daily mortality rate’, defined as ‘the number of chickens which have died in a house on the same day including those that have been culled either because of disease or because of other reasons, divided by the number of chickens present in the house on that day, multiplied by 100’. In other words, a certain percentage of chickens can be expected to die or to be culled because of disease or for other reasons, without that being any indication that the chickens are kept in what regulation 15 describes as ‘poor animal welfare conditions’.

Schedule 5A to the 2007 Regulations is concerned with the conditions in which conventionally reared meat chickens may be kept. Paragraphs 28 and 29 of Schedule 1 have a different focus. The mischief at which they are aimed is clear. Paragraph 28 prohibits the use of breeding procedures which themselves cause or are likely to cause suffering or injury to any of the animals concerned. Subject only to the relatively minor exception set out in sub-paragraph (2), the prohibition is absolute.

Paragraph 29, which is also under the heading of ‘Breeding procedures’, is not concerned with the breeding process itself, but with its consequences, i.e. the existence of any detrimental effect on health or welfare caused by the way in which farm animals have been bred. It recognises that breeders will wish to breed farm animals so as to increase productivity and that, while improvements in the quality of stock may have many advantages, selective breeding may also have adverse consequences for the health and welfare of the animals concerned. Accordingly Paragraph 29 deals with the keeping of farm animals which have been bred by genetic selection with a view to achieving particular characteristics, as compared with animals which have not been bred in this way – or, as most farm animals nowadays are the result of some breeding process, as compared with animals which have been bred without such adverse consequences to their health or welfare.

It can be seen that the wording of paragraphs 28 and 29 of the 2007 Regulations follows closely the terms of paragraphs 20 and 21 of the Annex to the Farming Directive. However, their structure is different. Paragraph 21 of the Annex to the Farming Directive provides that ‘No animal shall be kept … unless …’, while Paragraph 29 provides that ‘Animals may only be kept … if …’ However, the substance of the two paragraphs is identical as a matter of language and it would be extraordinary if the Secretary of State had intended when making the 2007 Regulations to introduce a substantive change to paragraph 21 of the Directive.

In my judgment, therefore, Paragraph 29 is correctly characterised as a prohibition which is subject to a proviso. The keeping of animals for farming purposes is prohibited unless it can reasonably be expected, on the basis of their genotype or phenotype, that they can be kept without any detrimental effect on their health and welfare. The structure of Paragraph 29 is clear. The language, and the terms of the Directive from which Paragraph 29 is derived, are not susceptible of any other interpretation.

In this respect, therefore, I do not agree with the view of the judge, who considered that Paragraph 29 could not be read in this way. His view was heavily influenced by the fact that breach of Paragraph 29 is a criminal offence and that there should not be an onus on the keeper of the animal to satisfy a criminal court (in practice the magistrates’ court) that as regards their genotype or phenotype the animals can be kept without detrimental effect on their health or welfare. However, that does not necessarily mean that in any criminal proceedings the onus would lie on the keeper to satisfy the no detriment proviso. We heard no argument about where the onus would lie and I do not wish to express any final view about it. However, the usual rule in criminal proceedings is that the burden of proof of all elements of the offence is on the prosecution and it seems to me, without the benefit of submissions, that Paragraph 29 is at least capable of being interpreted in this way.

Fast-growing chickens are animals kept for farming purposes. There is no other reason for keeping them. Accordingly, and leaving to one side the burden of proof, they may only be kept if (or they must not be kept unless) the no detriment proviso is satisfied. The real issue is what this proviso means.

The words ‘on the basis of their genotype or phenotype’ make clear that the prohibition is concerned with the adverse consequences of the way in which the animals in question have been bred. It requires a comparison with animals of the same species which have not been bred in the same way, and therefore do not have the same genotype or phenotype. It contemplates that the keeping of some breeds of animal for farming purposes may be prohibited because of the way in which they have been bred. Paragraph 29 is not concerned with the environmental conditions in which animals are kept which, in the case of conventionally reared meat chickens, are addressed in Schedule 5A.

I would therefore accept the Secretary of State’s case that Paragraph 29 is a prohibition on the keeping of farmed animals whose genotype and phenotype mean that, regardless of the conditions in which they are kept, they cannot be kept without detriment to their health or welfare. However, this needs some further explanation. There is a difference between detrimental characteristics which are inherent in the nature of the breed and which cannot be mitigated by changing the environmental conditions in which the animal is kept, and those which can be so mitigated.

For example, an animal which, because of the way in which it has been bred, is susceptible to cold will not suffer any detrimental effect on its health or welfare if it is kept indoors with appropriate heating during periods of cold weather. It can reasonably be expected, therefore, that such an animal can be kept without any detrimental effect on its health or welfare. The keeping of such an animal is not prohibited by Paragraph 29, although it may be that other provisions of the legislation would be engaged if the animal was kept outside in the winter, or if adequate heating was not provided. That might constitute a failure to take reasonable steps in all the circumstances to ensure that the needs of the animal were met, contrary to section 9 of the 2006 Act.

Coming closer to home, the conclusions contained in the RSPCA Report, if valid, also appear to illustrate this difference. If it is correct (and I emphasise ‘if’) that, when compared with slow-growing chickens, a particular breed of fast-growing chickens suffers from increased heart problems with consequential higher mortality, or leg development disorders because the chicken cannot support its own weight, it would seem likely (contrary to the Secretary of State’s view, although this is ultimately a matter for scientific evidence) that no improvement in the environmental conditions in which such chickens are kept could mitigate those detrimental effects upon their health or welfare. Those consequences would be inherent in the particular breed of chicken and the keeping of such a breed would be prohibited by Paragraph 29. The same might be true of the problems caused by hock burn and foot burn as a result of prolonged periods of inactivity. But that would depend on whether the inactivity was the result of stocking density (i.e. no room to move) which could be mitigated, or leg development disorder (i.e. not strong enough to stand) which could not.

The words ‘can reasonably be expected’ demonstrate that the test of expectation is objective. But the question arises, whose reasonable expectation is relevant and what information are they taken to know? The judge held that the standard is that of a reasonable person responsible for the animals in question since the obligation in Paragraph 29 is placed on such a person. So far, I agree. But the judge went on to say that such a person would not be taken to know the content of the RSPCA Report and the other scientific supporting evidence on which the appellant relies. In my judgment, however, this is a question of fact. The question is what a reasonable person keeping or intending to keep the animals in question for farming purposes would be expected to know. If they were a person responsible for a farmed animal, then they would have a duty under Regulation 4(2)(d) of the 2007 Regulations to have regard to the physiological and ethological needs of the animal ‘in accordance with good practice and scientific knowledge’, although what amounts to good practice and the state of scientific knowledge would be matters for evidence. In the case of meat chickens, it is not possible for this court to say whether such a person would be expected to be aware, for example, of the content of the RSPCA Report, or what view he or she would have about it – although it is fair to note that Mr Brown accepted in any event that the RSPCA Report was not conclusive. But it is reasonable to expect that a reasonable person keeping or intending to keep fast-growing chickens, who is after all responsible for their health and welfare, would at least be aware of the controversy about the keeping of such chickens and would take some steps to inform him or herself about it.

For the reasons already explained I would not accept that ‘can reasonably be expected’ requires any assumption that the animals will be kept in ordinary farming conditions. That is because Paragraph 29 makes no assumption about the conditions in which an animal will be kept. It is concerned with the characteristics of the breed, and with detriment which cannot be eliminated by improving the animal’s environment.

The judge glossed the words ‘any detrimental effect’ as meaning ‘any obvious or deleterious effect’. I respectfully disagree. A detriment may exist even if it is not obvious, so to say that Paragraph 29 is only concerned with obvious detrimental effects is not correct. Moreover, it is not clear whether the judge intended ‘deleterious’ to mean something different from, or more serious than, ‘detrimental’. If so, that would not be correct either. I would agree, however, that the more obvious any particular detrimental effect is, the more likely it is that the reasonable person will be aware of it.

The word ‘any’ was not included in paragraph 21 of the Annex to the Farming Directive, but was inserted into Paragraph 29 of the 2007 Regulations so that the concluding words read ‘that it can be kept without any detrimental effect on its health or welfare’. It was not suggested, however, that the insertion of the word ‘any’ caused any material change to the meaning of the paragraph and, again, it would be surprising if any such change were intended.

That said, I do not consider that the expression ‘any detrimental effect’ can be read literally. For example, in the same way as a breeding procedure likely to cause only minimal or momentary suffering or injury is not prohibited by paragraph 28 of Schedule 1 of the 2007 Regulations, I consider that it is implicit that the detrimental effect which Paragraph 29 is concerned to prevent must be more than minimal or momentary.

Similarly, it is inherent in the concept of reasonable expectation that, even if a detrimental effect occurs more often in animals of one particular breed than another, but nevertheless still occurs only in a tiny proportion of such animals, then it is not within the scope of Paragraph 29. In such a situation, it could reasonably be expected for the purpose of Paragraph 29 that animals of that breed can be kept without any detrimental effect on their health or welfare.

I am also inclined to think, although the parties made no submissions about this, that Paragraph 29 requires a holistic view to be taken of an animal’s health and welfare. Selective breeding may produce a breed which improves an animal’s health and welfare overall, albeit with some actual or potential detriment. If, taken as a whole, the health and welfare of the animals would be improved in this way, it is difficult to see why Paragraph 29 should be infringed.

It is important to be clear, however, that there is no question of balancing the advantage of higher productivity against the detrimental effect(s) on an animal’s health or welfare. If a given breed can reasonably be expected to suffer a detrimental effect on its health or welfare because of its genetic make-up, the prohibition on keeping that breed of animal applies regardless of any commercial advantages which it may have. Paragraph 29 unequivocally prioritises animal health and welfare over commercial benefit. But it does not prohibit selective breeding which can be undertaken to promote productivity with no detriment to an animal’s health or welfare.

For the reasons which I have sought to explain, I consider that there are some errors in the judge’s interpretation of Paragraph 29. In particular, as the appellant’s ground 2 suggests, the paragraph constitutes a prohibition which is subject to a proviso, such that a person is only permitted to keep an animal for farming purposes if the reasonable expectation proviso is satisfied.

However, this does not mean that the Code of Practice is unlawful. In R (A) v Secretary of State for the Home Department [2021] UKSC 37, [2021] 1 WLR 3931 the Supreme Court identified three types of case where a policy may be found to be unlawful by reason of what it says or omits to say about the law:

‘46. In broad terms, there are three types of case where a policy may be found to be unlawful by reason of what it says or omits to say about the law when giving guidance for others: (i) where the policy includes a positive statement of law which is wrong and which will induce a person who follows the policy to breach their legal duty in some way (ie the type of case under consideration in Gillick [1986] AC 112); (ii) where the authority which promulgates the policy does so pursuant to a duty to provide accurate advice about the law but fails to do so, either because of a misstatement of law or because of an omission to explain the legal position; and (iii) where the authority, even though not under a duty to issue a policy, decides to promulgate one and in doing so purports in the policy to provide a full account of the legal position but fails to achieve that, either because of a specific misstatement of the law or because of an omission which has the effect that, read as a whole, the policy presents a misleading picture of the true legal position.’

Similarly in R (BF) (Eritrea) v Secretary of State for the Home Department [2021] UKSC 38, [2021] 1 WLR 3967, in a judgment issued on the same day, the Supreme Court said:

‘52. Whenever a legal duty is imposed, there is always the possibility that it might be misunderstood or breached by the person subject to it. That is inherent in the nature of law, and the remedy is to have access to the courts to compel that person to act in accordance with their duty. An asylum seeker has the same right to apply to the courts as anyone else. Save in specific contexts of a kind discussed below and in our judgment in the A case, there is no obligation for a Minister or anyone else to issue policy guidance in an attempt to eliminate uncertainty in relation to the application of a stipulated legal rule. Any such obligation would be extremely far-reaching and difficult (if not impossible in many cases) to comply with. It would also conflict with fundamental features of the separation of powers. It would require Ministers to take action to amplify and to some degree restate rules laid down in legislation, whereas it is for Parliament to choose the rules which it wishes to have applied. And it would inevitably involve the courts in assessing whether Ministers had done so sufficiently, thereby requiring courts to intervene to an unprecedented degree in the area of legislative choice and to an unprecedented degree in the area of executive decision-making in terms of control of the administrative apparatus through the promulgation of policy.’

The appellant contends that the Secretary of State does not articulate the proper meaning of Paragraph 29 in the Code of Practice. That is true, but not because the Secretary of State has misinterpreted Paragraph 29. Rather it is true because the Code of Practice does not say anything about the meaning of Paragraph 29. It merely draws attention to it in the coloured boxes which I have mentioned, without any explanation of its meaning. The passages from A and BF (Eritrea) cited above make clear that the Secretary of State is under no duty to explain the meaning of Paragraph 29 in the Code of Practice.

In fact the only paragraphs of the Code of Practice which are specifically criticised are paragraphs 58 and 59, set out above, which deal with leg health. The appellant suggests that these paragraphs wrongly suggest that productivity considerations can be weighed against animal welfare detriment, such as to justify a detriment that otherwise violates Paragraph 29. If that were what these paragraphs suggest, it would clearly be wrong in law, but that is not how I read them. Rather they explain that leg disorders have many causes, of which genetics is only one, and make clear that welfare and health considerations should be taken into account when choosing a strain of chicken for a particular purpose or production system. The obvious intention is that a strain should be chosen which will not cause leg disorders, not that such disorders may be accepted for productivity reasons.

The appellant’s grounds of appeal contend that as a result of misdirecting himself as to the meaning of Paragraph 29, the Secretary of State has failed to undertake any prosecutions under that paragraph. It is true that there have been no such prosecutions, but the responsibility to prosecute is laid on local authorities by regulation 8 of the 2007 Regulations. The Secretary of State has power to direct that he, and not the local authority, should prosecute, but the appellant advanced no submissions to suggest that the non-exercise of this power is the result of any misdirection as to the meaning of Paragraph 29 or is otherwise unlawful. I therefore propose to say nothing further about this aspect of the case.

The appellant’s grounds of appeal also contend that the Secretary of State has failed to interpret paragraph 15(2) of Schedule 5A of the 2007 Regulations so as to give effect to Paragraph 29. Mr Brown said very little about this ground of appeal and I would reject it. Paragraph 15(2) of Schedule 5A is dealing with the environmental conditions addressed in the preceding paragraphs of the Schedule. It is not concerned with the adverse effects of selective breeding.

Although the grounds of appeal suggest that the thresholds set within the trigger system are unlawful, this point was not pursued. As the Secretary of State is currently consulting on possible revisions to the trigger system, and in circumstances where Mr Brown made no submissions about that system other than by reference to his proposed interpretation of Paragraph 29, it is unnecessary to consider this ground of appeal further.

Ground 5 was not pursued.

As I have explained, the judge decided as a matter of discretion that it was not appropriate to make a declaration as to the meaning of Paragraph 29 because this was an area where expert scientific judgment is required. The declaration which the appellant now seeks (see [38] above) is different from that which it sought before the judge. I agree, however, that this is not an appropriate case for a declaration in the terms now sought.

It seems to me that although in some respects I have taken a different view from the judge as to the meaning of Paragraph 29, the appellant’s claim for judicial review was a claim for two declarations, namely:

The judge dismissed that claim. On appeal, the appellant has hardly even contended that he was wrong to do so, recognising that this court cannot resolve the contested scientific issues. Instead the appellant has sought different relief which I would not be prepared to grant. It seems to me that in those circumstances the appeal should be dismissed.

I agree with Males LJ that the appeal should be dismissed for the reasons that he gives. I also agree with Males LJ’s analysis of paragraph 29 of Schedule 1 to the 2007 Regulations, together with the additional remarks of Underhill LJ.

There are some species of animals which plainly cannot be kept as farmed animals. For other species, such as chickens, the question of whether a particular breed can be kept as farmed animals will depend upon whether it can reasonably be expected that any characteristics of the breed, attributable to their genotype or phenotype and having an effect on their health or welfare (such as a susceptibility to a particular illness or type of injury, or a need to perform a particular type of behaviour), can be accommodated by the way in which they are kept.

To give the 2007 Regulations a sensible and purposive interpretation, that question has to be answered by looking at the well-being of the animal as a whole. It must also be permissible to disregard detrimental effects on health or welfare that are of entirely minimal effect or momentary duration, or which only occur in an insignificant proportion of the breed. But if, approaching matters in that way, there are detrimental effects on health or welfare that cannot be eliminated, then that particular breed cannot be kept as farmed animals, no matter how profitable or convenient the breed might be for a farmer. The 2007 Regulations do not permit the well-being of such a breed of animal to be traded off against the economic interests of farmers.

I agree that this appeal should be dismissed for the reasons given by Males LJ.  I also agree with him, however, that it is necessary for us to identify any respects in which the Judge erred in what he said about the meaning and effect of paragraph 29 of Schedule 1 to the 2007 Regulations.

As to that, I respectfully agree with Males LJ’s lucid exposition at paragraphs 40-61 above. But I would add one gloss on paragraph 58. I agree that in a case where a proportion, but only an insignificant proportion, of the animals of a particular breed will suffer harm (most obviously by developing some illness or injury) by being kept for farming purposes the prohibition in paragraph 29 should not apply. That conclusion may perhaps, as he says, be implicit in the phrase “can reasonably be expected”, but it can also be reached by declining to read the language of paragraph 29 literally: it is derived from an EU Directive, where the strict techniques of domestic drafting are not employed. In paragraph 57 Males LJ takes this approach in considering whether some minimum threshold of the degree and/or duration of suffering is implicit in the language of “detrimental effect”: I think a similar approach is appropriate in considering the degree of incidence of detrimental effects in the relevant population. The other point which I would note about paragraph 58 concerns Males LJ’s characterisation of the situation under consideration as one where the detrimental effect “occurs more often in animals of one particular breed than another”. That formulation reflects the fact that paragraph 29 is directed to characteristics which are peculiar to a particular breed – that is the effect of the phrase “on the basis of their genotype or phenotype” – from which it necessarily follows that there will be breeds which do not share that characteristic (or not in the same degree).