Advertising Watchdog Publishes Report on Tackling Harmful Racial and Ethnic Stereotyping in Ads

Photo by Yasin Yusuf

By Dr. Alexandros Antoniou, Lecturer in Media Law, University of Essex

On 3 February 2022, the UK’s regulator of advertising across all media, the Advertising Standards Authority (ASA), published its research in harmful racial and ethnic stereotyping in UK advertising. The survey highlighted a number of important issues that participant consumers raised about the depiction of people from different racial and ethnic backgrounds.

Ads that are likely to cause serious or widespread offence and/or harm owing to particular portrayals of race and ethnicity have long been regulated under the UK Code of Non-broadcast Advertising (CAP Code) and the Code of Broadcast Advertising (BCAP). Rule 4.1 of the CAP Code states that ‘Marketing communications must not contain anything that is likely to cause serious or widespread offence. Particular care must be taken to avoid causing offence on the grounds of age; disability; gender; gender reassignment; marriage and civil partnership; pregnancy and maternity; race; religion or belief; sex; and sexual orientation’. Equivalent provisions are found in Rule 4.2 of the BCAP Code. Marketers are urged to consider public sensitivities before using potentially offensive material and compliance is typically assessed with reference to several factors, including the context, medium, audience, type of product and generally accepted standards.

Advertising can play a role in legitimising stereotypes. Certain types of racial and ethnic stereotypes can, in particular, cause harm by creating a set of limiting beliefs about a person that might negatively affect how they perceive themselves, and how others see them. In the aftermath of the death of George Floyd (whose murder by a police officer in the US city of Minneapolis in 2020 sparked a global movement for racial justice and led to pressure for change across the world), the ASA has been reflecting on what further efforts could be made to address factors that contribute to Black, Asian and other minority racial or ethnic groups experiencing disproportionately adverse outcomes in different aspects of their lives.

As a first step, the regulator commissioned public opinion research in order to establish whether stereotypes associated with race and ethnicity can, when featured in ads, give rise to widespread or serious offence and/or contribute to real-world harm, such as unequal outcomes for different racial and ethnic groups. The research, which was conducted between March and June 2021, comprised two stages: a qualitative study that covered different interest groups, and a quantitative study that was designed to identify the extent to which attitudes and beliefs were held across individual communities and the UK as a whole. The research indicated that: ‘over half of Black, Asian and Minority Ethnic respondents felt that, when they were represented in ads, they are not accurately portrayed, and of those, just over a half felt people from their ethnic group are negatively stereotyped’.

Five categories of racial and ethnic stereotypes were identified by the research (some of which are interrelated):

  1. Roles and characteristics: overt or subtle stereotypical portrayals pertaining to appearance, behaviour, employment status, mannerisms, accent and preferences. Such portrayals may contribute to the homogenisation of vastly diverse groups and can be seen to reinforce or promote outdated views of a particular race or ethnic group.
  2. Culture: the exaggeration and mocking of accents, ‘lazy’ references to culture, cultural appropriation, and the use of imagery suggestive of colonialism.
  3. Religious beliefs and practices: repeated depictions of Muslim or Asian women wearing the hijab were seen by participants as ‘an easy stereotype that lacked authenticity’. There was, however, support for portrayals that did not draw specific attention to a person’s racial or ethnic background.
  4. Objectification and sexualisation: concerns were expressed about depictions of sexualised and/or objectified Black men and women as well as depictions that ‘fetishised and exoticised’ Asian women. However, positive portrayals of the diversity of body shapes and sizes were generally welcomed.
  5. Use of humour at the expense of other ethnic groups: making fun of a group or their appearance, culture or tastes, e.g., the use of different accents can be seen as mocking or ‘othering’ by reinforcing the idea that people from racial or ethnic minorities who speak with an accent are different from White or Western people.

Moreover, the research highlighted three potential types of harm that could develop from adverse portrayals of race and ethnicity:

  1. reinforcement of existing stereotypes through the repeated use of certain portrayals (often described as ‘always showing us the same way’, e.g., the casting of Asian men as shop keepers, waiters and taxi drivers or subtle reinforcements of a servile role). The perceived harm in relation to this was seen in making it easier for others to see people from racial or ethnic minorities as different to the mainstream (‘othering’);
  2. the emergence of new tropes which continue creating a one-dimensional picture of Black, Asian and other minority racial or ethnic groups; and
  3. perpetuating or implicitly reinforcing racist attitudes by depicting racist behaviour: such depictions were felt to pose a risk of evoking past trauma and reinforcing prejudice (even where it was understood that the advertiser’s intention was to challenge negative stereotypes within the messaging of the ad).

The research did not give the ASA reason to believe that its interpretation and application of the Codes’ rules were generally out of step with consumers’ and stakeholders’ opinions. The findings can, however, bring more clarity and valuable insights on the types of ads that pose a risk of causing harm and/or offence. At the end of 2022, the regulator will conduct a review of its rulings in this area to identify newly emerging areas of concern and ensure that it is ‘drawing the line in the right place’.

At this stage, it is not anticipated that a new targeted rule will be introduced into the Advertising Codes to ban the kinds of portrayals identified in the report. Nevertheless, the Committee of Advertising Practice (CAP) and the Broadcast Committee of Advertising Practice (BCAP), which are responsible for writing and updating the UK Advertising Codes, will consider whether specific guidance on racial and ethnic stereotypes is necessary to encourage creative treatments that challenge or reject problematic stereotypes and diminish issues arising from the repeated presentation of a specific race or ethnicity in a particular way. Finally, the research findings will be presented to industry stakeholders and training will be offered to support advertisers where necessary.


This article was first published on the IRIS Merlin legal database and is reproduced on the ELR Blog with permission and thanks. The original piece can be viewed here.

Nazanin Zaghari Ratcliffe: Her Journey to Freedom and the Lessons We Can Learn

Photo of Richard Ratcliffe at the candlelit vigil outside the Foreign Office on Friday 05 Nov. 2021, eleven days into his hunger strike, via Flickr.

Professor Carla Ferstman is a lawyer and an activist. Before joining the School of Law in 2018, she directed REDRESS, an organization dedicated to helping torture survivors in all parts of the world to seek justice for all the harm they suffered. That is where she first met Richard Ratcliffe, the husband of Nazanin Zaghari Ratcliffe who was released earlier this week on 15 March 2022 after almost six years of being kept as a hostage in Iran.

Kate Clayton, Senior Communications Officer at the University of Essex, spoke to Carla to find out more about her perspective on Nazanin’s journey to freedom and to ask her what lessons we might be able to draw from her case.

Why has the plight of Nazanin and her family resonated with so many people in the UK and beyond?

On a human level, it is hard to fathom what it must feel like to have one’s family torn apart by such an arbitrary, brutal act and to feel so powerless over so many years. So this was about compassion first of all. But also, Richard’s advocacy, his unwillingness to be quieted in the face of the injustice he and Nazanin faced helped to bring and keep people on board.

Do you think there were any turning points in the campaign?

Yes, several, and I will focus on the positive ones.

First, is the recognition by the UN Working Group on Arbitrary Detention that Nazanin’s detention was arbitrary, that she was likely to have been arrested because of her status as a dual Iranian-British national, and that she should be immediately released. This 2016 decision made it clear that this was no ordinary criminal case where the UK should sit back and wait for justice to take its course. No, Nazanin was being targeted. Removing the veneer of a criminal justice justification for her detention was really important because it helped to move the UK Government towards a position where it understood that it had to act. Passivity was not an option.

Second, was the coming together of many of the families of detainees, mainly dual nationals and Iranian nationals with foreign links. This was crucial to change the narrative about what was happening. All the stories were so similar – this was a form of hostage-taking. It was also important to counter isolation and build a sense of common solidarity. 

Third, was the 2019 decision by then Foreign Secretary Jeremy Hunt to grant Nazanin diplomatic protection, meaning that the UK Government had recognised formally that the harm caused to Nazanin was a harm to the UK Government and one for which it could intervene as a state to state claim. This was a landmark recognition.

Fourth, was the March 2022 repayment by the UK of a £400m debt that had been outstanding since the 1970s in relation to an outstanding order for military equipment.

You managed to involve Essex students in the campaign. How did this go?

Iran’s human rights record was being considered by the UN Human Rights Council in Geneva through its universal periodic review process. The students, under the auspices of the Human Rights Centre Clinic helped prepare a submission on behalf of seven families to highlight the injustice of their situation. This came at a really important time and was a start of much more robust joint advocacy by families of detainees. This was quite a unique opportunity for the students to work on such a concrete, live case involving real people undergoing serious human rights violations in real-time.

Have campaigns like this impacted your academic research?

Indeed, I just recently co-authored with my colleague Dr Marina Sharpe a journal article which considers whether the arbitrary detention of dual and foreign nationals in Iran violates the Convention on the Taking of Hostages and may constitute a crime against humanity. We hope this will be useful to ongoing scholarly debates and also assist the many organisations who are following these issues and governments whose citizens continue to be affected by the practice.

What do you think comes next for Nazanin and Richard?

One of the wonderful things about the freedom they now have is that it is absolutely for them to figure out their next steps. Something we may all take for granted, I imagine for them feels very luxurious.

What next for the other cases?

There are still so many people who remain arbitrarily detained in Iran in a hostage context, many for multiple years, under very difficult circumstances. And the practice is also happening in more and more countries. The advocacy must continue until the practice stops.


This piece was first published on the Blog of the University of Essex and is reproduced on the ELR Blog with permission and thanks.

The Case of the Caterpillar Cakes: Why Legal Protection for a Shape is So Hard to Come By

A very hungry caterpillar (image via Flickr)

By Prof. Stavroula Karapapa and Dr. Alexandros Antoniou, School of Law, University of Essex

UK retailers Marks and Spencer (M&S) and Aldi have finally called a truce to the trademark-based legal spat pitting their caterpillar cakes, Colin and Cuthbert, against each other. While details of the settlement have not been made public, Aldi’s Cuthbert will not reportedly return in quite the same form. As Aldi tweeted, upon announcing that an agreement had been reached:

Colin the Caterpillar, a roll sponge cake decorated with milk and white chocolate icing and sprinkles, has been an M&S stalwart for more than 30 years. The retailer claims to have sold more than 15 million to date. Since 2011, rival products with similarly alliterative names have appeared: Asda’s Clyde, Tesco’s Curly, Waitrose’s Cecil, Co-op’s Curious and Cuthbert.

In April 2021, presumably because it was thought that, of all the caterpillar cakes, Cuthbert most closely resembled Colin, M&S launched legal proceedings to protect its intellectual property and get Aldi to remove the product from its shelves. The retailer claimed that the similarity between the two cakes would lead consumers to think that they were of the same standard, thereby allowing Cuthbert to ride on Colin’s coattails.

Companies often seek to protect, as trademarks, certain signs that help them distinguish their products and services from those of their competitors, such as brand names, logos and slogans. Along with patents (which protect innovative technical solutions) and copyright (which protects creative or intellectual works such as books and music), they are a form of intellectual property.

The Colin v Cuthbert dispute hinges specifically on trademarks, because it related to the distinctive characteristics of commercial assets. M&S has held trademarks in the UK in relation to Colin’s name and green packaging since 2009 and 2020 respectively, but these have not been infringed.

The problem for M&S is that its trademark would probably not extend to the underlying idea of a chocolate roll with a smiley face on it. Securing protection for the shape of a product is actually quite difficult in trademark law – not to mention proving that a competitor has presented his goods as those of somebody else.

The four-finger-shaped KitKat chocolate bar and the shape of the London taxi are two examples of iconic shapes that have not succeeded. One key reason behind this is that the average consumer doesn’t usually make assumptions about the origin of products on the basis of their shape, or that of their packaging, when other graphic or word elements are absent.

Caterpillar cakes have seen decades of children grow up (image via Flickr)

Passing Off

There have been cases in the UK where shapes and packaging have attracted protection under what jurists refer to as the law of passing off. Passing off offers legal protection against harm to what is termed the “goodwill” of a business. And it can be used to protect unregistered trademarks.

Goodwill here is a legal concept, which refers to a business’s means of attracting people’s custom. It is harmed when a trader suggests – through misdescription on packaging or parasitic copying of a well-known product – that their product or service has some association or connection with another trader, when this is not the case.

In 2015, pop-star Rihanna famously won her battle to stop fashion brand Topshop using an unlicensed image of her on a T-shirt. She did so by convincing a judge that customers buying the top would think she had endorsed it. The Court of Appeal ruled that the unauthorised use of her photograph amounted to passing off.

Passing-off claims are notoriously difficult to establish. If someone promotes their bottled drinking water business as “the De Beers of still water”, they may be infringing the De Beers trademark but it is unlikely that a judge would find that they were passing themselves off as connected to De Beers in any commercial sense.

To win a passing-off claim against Aldi, M&S would have essentially had to prove that Colin had built up such a reputation as a signature product – in the same way that Rihanna had done as a music artist and style icon – that customers would be able to recognise it without difficulty. The retailer would also have had to prove that Cuthbert was so similar to Colin that consumers, after opening the packaging, would be misled into thinking that the two were somehow associated.

What is more, the fact that so many supermarket chains now have their own version of a caterpillar cake – coming in a variety of sizes and decorative features – would not have helped in establishing that Colin is unique. M&S would have had to prove that in the minds of cake buyers in the UK, the caterpillar has not become a generic shape for cakes.

Further, Aldi’s excellent Twitter campaign – and the widespread publicity the lawsuit has attracted – will have also contributed to dispersing any consumer confusion.

Lastly, it would have been difficult for M&S to argue that Aldi’s caterpillar cake had damaged or had the potential to damage the goodwill in Colin – that is, its power to attract and retain buyers of the cake. The market for caterpillar cakes is saturated: there are just so many to choose from.

The agreed settlement between the two parties means there was no court judgement on the facts. The terms of the settlement remain confidential. It is unknown whether liability – that is, blame – was admitted by Aldi for the wrong allegedly suffered by M&S.

Rarely does a party in a settlement walk away thinking, I have won. Typically, there is no winner and no loser in a negotiated resolution. This case will nonetheless have seen both M&S and Aldi benefit from the kind of marketing exposure that money cannot readily buy.


This article was first published on The Conversation and is reproduced on the ELR Blog under a Creative Commons Licence. View the original article here.

The Future of AI Liability in Europe

Image by VectorStock

Artificial Intelligence (AI) could revolutionise the world-wide economy as well as the way we live, work and interact with each other. While this new technology certainly presents great potential, it also comes with important risks to human life, health and wellbeing – among other risks.

In an effort to prepare for this new environment, the European Commission has been at the forefront of several initiatives that aim to provide a harmonised regulatory framework for the safe deployment and use of AI systems across Member States [1]. Amongst its most recent initiatives is a public consultation on how to adapt civil liability rules to the digital age and artificial intelligence. This public consultation, which closed on 10 January 2022, aimed to collect views on:

  1. how to improve the applicability of the Product Liability Directive (PLD) to the digital age, including AI, and
  2. whether there is a need to further harmonise rules of liability for damage caused by AI systems beyond the PLD.

The consultation is an important first step towards building a robust liability framework fit to address the current and future challenges posed by AI and the digital age in Europe. The changes that could be implemented as a result of the consultation could be immense and produce far-reaching consequences. Understandably, this public consultation attracted a high level of interest from various stakeholders, including businesses (Google, Bosh, Siemens, Avast), consumer organisations (BEUC, France Assos Santé), insurers (AXA, Insurance Europe, France Assureurs), NGOs, interest groups, legal scholars as well as members of the general public. In total, the European Commission received around 300 responses.

Pr. Jonas Knetsch (University of Paris 1 Panthéon-Sorbonne) and Dr. Emmanuelle Lemaire (University of Essex), assembled a small ad hoc research group, comprised of Pr. Michel Cannarsa (The Catholic University of Lyon), Dr. Laurie Friant (University of Paris 1 Panthéon-Sorbonne) and Pr. Simon Taylor (Paris Nanterre University), to produce a report in response to the consultation.

Overall, the authors of this report were of the view that the PLD should be adapted to enhance consumer protection in the digital age and increase legal certainty for all stakeholders. The authors also recognised that AI technology posed specific challenges and recommended that complementary measures be adopted to ensure the safe deployment and use of AI systems across Member States.

Adapting the PLD rules to the digital age and AI

The Product Liability Directive, which came into force on 30 July 1985, was a response to the increasing demand for consumer protection in a hyper-industrialised environment where goods were mass-produced, and mass-consumed. In essence, the Directive aimed to offer a high level of protection to consumers while ensuring that producers did not bear an undue burden. It was thus designed to strike a careful balance between the interests of both consumers and producers.

Yet, we must remember that the Directive was implemented at a time when the Internet was still in its early days, the use of AI remained largely theoretical, marketplaces were positioned in the ‘physical world’, and concepts such as ‘circular economy’ and ‘the Internet of Things’ (IoT) were simply non-existent. To say that the PLD – which did not undergo any major changes since 1985 – is in need of reform is certainly an understatement.

In order to adequately adapt the PLD to the digital age and AI, the authors of the aforementioned report took the view that the scope of application of the PLD should be extended, and in particular that:

  • the concept of ‘product’ should be expressly extended to intangible goods,
  • the concept of ‘producer’ should be extended to include online marketplaces and remanufacturers,
  • the concept of ‘damage’ should be extended to include specific types of immaterial loss (i.e. privacy or data protection infringements not already covered under the General Data Protection Regulation, and damage to, or the destruction of, data).

The authors of the report also recommended the amendment of specific PLD rules in certain situations, and more specifically:

  • the suppression of the development risk defence for AI products only,
  • the suppression of the 10-year longstop period in case of death or personal injury,
  • a clarification of the conditions enabling the 3-year limitation period to start running,
  • an alleviation of the burden of proof of ‘defect’ and ‘causation’ for products classified as ‘technically complex’ (which would include AI products and The Internet of Things).

In addition to recommending that the PLD be adapted, the authors of the report were also in favour of the European Commission adopting complementary measures in the context of AI to account for the specific features presented by this technology (autonomy, complexity, opacity, vulnerability, and openness).

Adopting complementary measures in the context of AI

The regulation of AI is proving challenging across legal systems, not least because of the difficulty in defining what AI is and what can be classified as an AI system. The European Commission made a recent effort to try and offer a clear – but open – definition of the term ‘AI system’ to ensure legal certainty while providing the necessary flexibility to accommodate any future technological developments. As the definition currently stands, an AI system means software that is developed with some specific listed techniques and approaches ‘and can, for a given set of human-defined objectives, generate outputs such as content, predictions, recommendations, or decisions influencing the environments they interact with.’[2] The definition is quite broad, and in consequence, the range of products based on – or using –  AI systems can be diverse and include voice assistants, image analysing software, search engines, speech and face recognition systems, as well as advanced robots, autonomous cars, drones or Internet of Things applications. Not all these products present the same type or level of risk, and some AI-based products are therefore more dangerous than others.

The authors of the report recommended that the European Commission consider:

  • the harmonisation of strict liability where AI-based products or services create a ‘serious risk of damage’ to consumers with an option to allow Member States to offer more protective liability rules to consumers,
  • the harmonisation of mandatory liability insurance for certain AI products,
  • the harmonisation of liability rules regarding the compensation of specific types of immaterial loss beyond the PLD (i.e. privacy or data protection infringements not already covered under the General Data Protection Regulation, and damage to, or the destruction of, data).

If you are interested in knowing more about the recommendations made by this university group to the European Commission, you can find a copy of their report (no. F2771740) – written in French – on the EC website or download it directly from our blog below:


[1] See e.g. European Commission, Communication from the Commission to the European Parliament, the European Council, the Council, the European Economic and Social Committee of the Regions – Artificial Intelligence for Europe (COM(2018) 237 final); European Commission, White Paper on Artificial Intelligence – A European approach to excellence and trust, (COM(2020) 65 final); European Commission, Communication Coordinated Plan on Artificial Intelligence (COM(2021) 205 final); European Commission, Proposal for a Regulation of the European Parliament and of the Council laying down harmonised rules on Artificial Intelligence (Artificial Intelligence Act) and amending certain Union legislative Acts (COM(2021) 206 final).

[2] European Commission, Proposal for a Regulation of the European Parliament and of the Council laying down harmonised rules on Artificial Intelligence (Artificial Intelligence Act) and amending certain Union legislative Acts (COM(2021) 206 final), Article 3(1).