Opinion: Lab-grown ‘Clean meat’ is a misnomer, in more ways than one

Mick Keogh, Australian Farm Institute, 13/03/2018

Producers of laboratory-cultured animal cells for food needs to come up with a description other than ‘meat’, argues Australian Farm Institute’s Mick Keogh


SO-CALLED ‘clean meat’ is the latest darling of the stock market in the US, with major investors putting their money into start-ups like Impossible Foods, Hampton Creek, and MosaMeat.

Quite a few commentators and activists have claimed that these developments in alternative proteins signal the end of animal agriculture as we know it, citing ‘significant advantages’ in environmental and animal welfare outcomes as key reasons for the transition.

Lab-grown meat burger

Two issues arising from the development of ‘clean meat’ need careful consideration. The first may seem trivial but in fact is of major importance: that is what these products are called. The second, equally important, is whether they can justify claims to be more environmentally friendly than the foods they aim to replace.

Meat is a term that has long-established meaning, whether considered formally via dictionary definitions, or – more importantly – in the minds of consumers. The Oxford Dictionary states that meat means the flesh of an animal, typically a mammal or bird, as food. Clearly, a chemically reconstituted artificial protein that has been produced in a factory utilising multiple chemical additives, some of which are derived from genetically modified animal and plant matter, does not in any way fit the above definitions. Therefore, it is quite incorrect to use the term ‘meat’ as a name for these products.

The issue of cultured or ‘lab grown’ meat is slightly different, in that it is derived from animal cells multiplied via culturing in a laboratory. It could be argued that these products can be described as ‘meat’, although consumers may be somewhat surprised if they were provided with the complete details of how these products are created.

Leaving cultured meat aside, the use of the term ‘meat’ to describe chemically reconstituted artificial protein is not only misleading but is also an attempt to appropriate the value that consumers attribute to products referred to as meat – and are prepared to pay for – to create a market for these new products.

The importance of food terminology and names has been highlighted in the past, with the names of wine regions in France being a classic example. The European Commission moved to enact Geographical Indicator regulations some time ago, to protect wine producers in regions like Champagne and Burgundy from having their brand equity eroded through the use of these terms to describe wine from other regions.

“There is a very strong argument that livestock producers and meat processors should have available similar protection”

The EC maintains that naming things correctly “matters economically and culturally”. There is a very strong argument that livestock producers and meat processors should have available similar protection, something which is currently a subject of debate in the USA.

The second issue of dispute is whether these products can justify the claim that they are somehow environmentally ‘cleaner’ than actual meat.

The websites of some of these companies are littered with claims about the environmental damage associated with conventional livestock production, but many of the so-called facts they cite are at best contentious, and at worst blatantly and obviously incorrect.

The following quotes from the website of Impossible Foods (manufacturer of an entirely plant-based hamburger pattie) are classic examples:

  • Because we use 0% cows, the Impossible Burger uses a fraction of the Earth’s natural resources. Compared to cows, the Impossible Burger uses 95% less land, 74% less water, and creates 87% less greenhouse gas emissions.
  • The way the world produces meat today is taking an enormous toll on our planet. According to livestock researchers, animal agriculture uses 30% of all land, over 25% of all freshwater on Earth, and creates as much greenhouse gas emissions as all of the world’s cars, trucks, trains, ships, and airplanes combined.

These claims are not referenced to any authoritative research and appear to rely on some heavily criticised life cycle assessments published some time ago which contained fundamental errors.

Much mis-quoted water use

One example was the much quoted “200,000 litres of water to produce one kilo of beef”. This and other similarly outlandish figures were arrived at by making patently incorrect assumptions, such as that all the rain that falls on pastures being grazed by cattle should be attributed to those cattle, and any runoff, deep soil drainage or evaporation simply ignored! Detailed, objective life cycle assessment of Australian beef cattle production systems has identified that actual water use ranges between 100 and 300 litres per kilogram liveweight, equal to between 200 and 600 litres per kilogram of meat. Exactly what objective data Impossible Foods has utilised to make the above claim is unknown, but it seems to be highly misleading.

Similar criticisms apply to the comparison of livestock and transport emissions. The above appears to be based on a heavily criticised United Nations report released in 2006 entitled “Livestock’s Long Shadow”. This report compared greenhouse gas emissions estimated for a very broadly defined livestock sector (including land clearing, fertiliser, farm, feed, transport and processing) with just the emissions arising from fuel use in transport, ignoring all the emissions associated with vehicle manufacturing, tyres, and the construction and maintenance of roads, bridges, railways and airports. This was blatantly an “apples vs oranges” comparison, which one of the report’s authors conceded in 2010.

In the end it is up to consumers to decide whether they prefer meat or non-conventional meat substitutes, but they should at the very least be presented with objective information to enable informed decision making, and this includes the naming of these products.

Therein lies a problem. The use of the term ‘clean meat’ is obviously misleading on both counts, as the products have questionable ‘clean’ credentials and are very obviously not meat.

Exactly what chemically reconstituted artificial proteins should be called is yet to be agreed – but perhaps somebody will eventually come up with an acronym!


  • See James Nason’s earlier treatment of this topic here.


Your email address will not be published. Required fields are marked *

Your comment will not appear until it has been moderated.
Contributions that contravene our Comments Policy will not be published.


  1. Deborah Jobson, 14/03/2018

    Personally, I find it scary that a GM engineered product called “clean meat” can in any way be associated with or labelled as a “meat”. Many people reject GM products but they need to understand the product contains GM material; others who don’t understand the effects of GM foods will purchase solely based on price and taste.

  2. Dan Smith, 14/03/2018

    The issue of food terminology and names doesn’t seem to have applied in relation to dairy, with all sorts of nut and bean concoctions seemingly able to use the term ‘milk’.

  3. Paul D. Butler, 13/03/2018

    It is a real tragedy that folks in the beef business continue to propagate the terminology deception……..include most of those who write about these products. Quit calling this stuff “meat”……it is NOT meat in any way, shape or form. Call it for what it is……..Lab Slime. Champion truth in labeling………NOT deception. Call it “Pure Lab Slime” if you prefer…….just use accurate terminology.

  4. Ian McCamley, 13/03/2018

    Well said Mick Keogh. About time someone took these people on for trying to steal the mighty name ‘meat’, and for backing their artificial substitute with ‘artificial facts’. Its a shame the artificial meat substitute people don’t understand just what a special privilege it is to manage animals to sustainably turn the worlds abundant grass, inedible to humans, into genuine truly satisfying meat.

  5. Glenn Vassallo, 13/03/2018

    The inability of cultured meat to reproduce the complexity of a piece of meat is the reason that those of us in the livestock industry have little to worry about. Cultured meat will find its way into various food products, but just as synthetic fibre did not replace wool and cotton, cultured meat is unlikely to replace the real thing. As an example, given a choice, would the average consumer prefer a wool jumper or a polyester jumper? So long as care is taken from an ethical standpoint, and productivity can continue to be improved, then I suspect the same story will play out for meat.

  6. Sandi Jephcott, 13/03/2018

    This is a comprehensive article but it does not cover what the beef industry the world over is doing about the issues of dispute and the misnomers that are being quoted. Also how will the price of the ‘pink slime’ compare to beef

  7. Paul Franks, 13/03/2018

    I think without a doubt should artificially created meat-like products be created for an affordable price, it will signal the end of the beef industry in Australia as all the major markets switch over.

    All the major burger outlets would quickly use emotive marketing to try to gain an edge. Just look at how the mulesing issue played out.

    The key word in your comment is ‘meat-like,’ Paul. From what we understand, lab-grown meat has none of the texture or structure of a steak, or even a sample of mince. It is like a fine slurry, made up of individual muscle cells. Dare we use the phrase, ‘Pink slime?’ Editor

Get Beef Central's news headlines emailed to you -