Agricultural Challenges: Ending Fantasies, Returning to Realities

The relentless decline in the relative weight of agriculture

There is a growing number of studies on agriculture using diverse approaches, whose compatibility is not always considered. We also observe an increasing number of field experiences that various actors in the agricultural world can have, in many countries, and in very different pedoclimatic, sociotechnical, and territorial contexts. Their confrontation leads to the general conclusion that we can hardly speak of the challenges, stakes, and means of "global agriculture" and that we must today accept the plurality of agricultures at different levels, on different registers, and from different theoretical and practical angles. For the approach to the agricultural world is imbued with ideologies, in the sense of Canguilhem, meaning conceptions that place the search for truth under the dependency of value judgments. Understanding the real agricultural constraints (physical, biological, ecological, human, sociotechnical, territorial, geopolitical, etc.) becomes the first step to move away from discourses on an imaginary agriculture, disconnected from the concrete realities of agricultural production. Agriculture is primarily a production of food products, or more broadly energy or material products, and a real improvement must be such that the return on energy investment is always higher, without compromising any of the supports of agriculture, which are soil, climate, productive life, and the entire associated biosphere, but also the humans directly involved in this productive work. The excessive energy cost of AI suggests being very cautious regarding it, at least concerning all agricultures.

"Understanding the real agricultural constraints becomes the first step to move away from discourses on an imaginary agriculture,

disconnected from the concrete realities of agricultural production."

In fact, there is clearly a first major challenge, the most important, limiting factor for any other objective, which is to feed 10 billion people[1] in about 30 years, around 2050. Demographic projections are clear. Yet, agricultural land use must be limited as much as possible, both in terms of area, greenhouse gas production, maintaining biodiversity, use of pesticides, energy inputs, etc., sustainably, meaning without compromising production capacities for decades, even centuries to come, no country can claim to have an agriculture capable of "Feeding the World".

The FAO goes in the same direction. Even countries with a potential far superior to that of France cannot have this objective. Let us understand that even if one or more countries had the capacity through their food surpluses to play an important role in feeding the world, to the point of wielding "the food weapon", it would now be an illusion because the global economy is too integrated and the main problem lies at the level of "means of payment". We can easily cite a few countries whose agricultural potential far exceeds their domestic needs. For example, Brazil, the United States of North America, Russia, Ukraine, but also Oceania and almost all Spanish-speaking America. Other important countries are structurally net importers, such as Japan, the United Arab Emirates, Saudi Arabia, Egypt, the United Kingdom, which leads structurally to significant trade in raw agricultural products, but also in finished food products. However, the history of the world, since the beginning of agriculture, shows that in case of food production shortages, it is the agricultural world itself that suffers the most, especially as it is poorer.

It is indeed thanks to agriculture, the worst mistake of humanity according to Jared Diamond (1987), that civilizations have flourished, supported by non-farmers, who are in the minority, at the expense of farmers, who are in the majority. The question today becomes how to feed 10 billion people who will increasingly be predominantly urban dwellers, by an ever smaller minority of farmers whose level and quality of life must be maintained, otherwise the very foundation of current human civilizations will be threatened.

No country can satisfy all the needs or demands of consumers, for pedoclimatic reasons. Coffee, tea, cocoa, natural vanilla, pepper, to give five well-known archetypal examples, while having acquired the status of universally consumed products, can only be produced by a limited number of countries. Many more basic products, both plant and animal, are gradually experiencing the same fate. In fact, we see that the relatively low cost of freight, for the moment, can lead to significant and growing exchanges of products for reasons of competitiveness. At the same time, while the number of people fed per farmer continues to grow, agriculture weighs less and less both in the overall value of traded products and in the overall value of productions. There is a little-studied paradox in anthropological terms: without relative enrichment of farmers, can this situation be sustainable? Everything happens as if agriculture, and therefore farmers, count less and less, regardless of the angle of analysis. We are well aware of the economic and financial reasoning that "explain" this situation and this evolution, but can this last reasonably?

The dependence on fossil energies

Joseph Klatzmann, in 1975, had already recognized that a planetary population of 10 billion people should be reached during the 21st century, and that it was necessary to analyze whether it was possible to feed such a population. In 1700, the world population was slightly below 700 million, close to a billion in 1800; in 1900, it reached 1.6 billion; in 1975 about 3.6 billion, and in 2024, 8.2 billion. According to average projections, it should approach 10 billion by 2050 before inexorably declining.

This acceleration – which has been slowing down slowly for 50 years – has occurred because farmers around the world were able to meet the necessary production growth, since the number of undernourished people, estimated at more than 950 million in the 1950s (for a population of 2.5 billion, or 38%) has decreased to less than 733 million today (for a population of 8.1 billion, or just over 9%) and the year 2025 seems set to be a record year for global agricultural production. Is this decrease from 38% to about 9% evenly distributed between the agricultural world and the urban world? Such analyses have not been made. The FAO's SOFI 2024 study compares the urban world (55.6% globally) and the rural world (44.4%), knowing that this separation does not have the same meaning depending on the development of countries worldwide. Whatever happens, farmers are globally a minority, even in rural areas, but even less so as countries are poorer. We cannot, in this article, analyze finely the relationship between "hunger", development, and agricultural population. It remains that it is in the least urbanized countries that hunger remains significant, and it affects rural areas more than urban areas. Nevertheless, there appears to be a sort of residual "hunger" in urban areas, even developed ones, which has a different meaning. The current global world of abundance is built with an increasingly declining population of farmers who produce to feed an ever-growing urban population.

"The current global world of abundance is built with an increasingly declining population of farmers

who produce to feed an ever-growing urban population."

Joseph Klatzmann also identified how agriculture could reach such a level of production: through nitrogen fertilizers (and so-called base fertilizers), phytosanitary products, genetic selection, and mechanization. He already pointed out the necessary dependence on fossil energies and the environmental risks. Harchaoui and Chatzimpiros (2018), based on the French situation from 1882 to 2016, showed the radical transformation that allowed French agricultural production to multiply by more than three, with an ROEI (return on energy investment) multiplied by 2.33, but with the consequence of a radical dependence on fossil energies. In fact, the level of global agricultural production is now dependent, for at least 50% of its production volume, on inputs derived from fossil energies (mainly oil and gas). The necessary global growth of agricultural production in 2050 will be even more dependent on fossil energies, unless there is a radical change in production and final consumption paradigms. The production of about 113 million tons of nitrogen reduced through the Haber-Bosch process in 2023 requires about 170-180 million tons of oil equivalent. The energy cost of other production factors (other fertilizers, phytosanitary products, machines) leads to a need for about 300 million tons of oil equivalent to enable current global agricultural production, which is about 2% of total global energy consumption.

However, the Haber-Bosch process necessarily uses fossil energies, unless there is a technological breakthrough, as do phytosanitary products, the extraction and processing of other fertilizers, and of course the majority of agricultural machinery. In general, agricultural needs for fossil energies represent about 3% of the global total, and shifting from fossil energy to decarbonized energy requires too much technical innovation to be credible in the medium term. This is relatively low, but it is absolutely essential, as it is mainly about human food. In their absence, primary agricultural production would decrease by at least about 50%. In the case of only a downward trend, that is to say, of rising prices, what would happen?

The current drop in fossil energy prices, the result of political choices related to recent events, will certainly not be sustainable, as the laws of physics are relentless. Anticipating a decrease in agricultural dependence on inputs derived from fossil energy is therefore an integral part of the challenges of global agricultures, knowing that the prices of cereals and sugar are now directly linked to the price of oil due to their possible substitution (biofuels).

The impact of agricultural production on the environment

The ability to meet global food demand must be acknowledged and recognized, as it has resulted from a multitude of actions in almost every country on the planet, but it has nevertheless had a consequence, certainly predictable (cf. again Klatzmann), but very difficult to avoid, that of a tremendously increased pressure on land spaces, but also maritime ones (fishing has also experienced strong growth, albeit on a plateau, compensated by aquaculture). This pressure has exacerbated the decline, already longstanding since the beginning of the Holocene, of biodiversity, and also a decrease in soil quality, which is a threat to future productions.

Despite the continuous growth of yields from 1946 to 1990 practically everywhere in the world, but stabilized since, deforestation and the increase in agricultural pressure have continued, especially as migration from rural areas to cities has continued in all countries, at speeds that vary according to countries, but overall accelerating during this period. Most major cities are almost all located, for obvious historical reasons, in regions initially very conducive to agriculture. Their expansion leads to considerable losses of good agricultural land.

Moreover, agricultural development in countries whose enrichment modifies consumption habits is oriented towards: more animal products, more processed products. The first aspect increases the necessary area and greenhouse gas emissions, the second aspect shifts industrial activities and leads to increased resource and energy needs and recycling needs (costly in energy). This issue of the necessary increase in produced volumes (at the global level) leads to difficult debates about the place of the "wild" environment, which is in the process of disappearing. This debate, exposed under the Anglo-Saxon terminology "land sharing versus land sparing", is far from being closed. Because the FAO's recommendations are to globally increase production by 50% by 2050, knowing that the expansion of cultivated areas now seems very limited.

Of course, agriculture is far from being solely responsible for the decline in biodiversity, greenhouse gas emissions, environmental pollution, forest fires, and land artificialization. It remains that urban populations now increasingly exceed rural populations, the gap is widening every day between the agricultural world and the urban world, and the loss of contact between urban dwellers and agriculture leads to demands from urban dwellers that are out of sync with the problems that farmers must solve. The tensions between agricultural and urban worlds can only grow.

"The gap is widening every day between the agricultural world and the urban world, and the loss of contact between urban dwellers and agriculture leads to demands

from urban dwellers that are out of sync with the problems that farmers must solve."

The Kaya equation, originally written for CO2 emissions, is also adapted for ecological footprints, pollution, and the main planetary limits now well known. The multiplicative factor between GDP/ha and population, reduced to the corresponding area, shows the "physical" limits of any productive system. It is likely that one of the corresponding challenges will be to make urban dwellers understand that their food consumption patterns must change in order to demand less agricultural land. As Jean-Marc Jancovici has noted many times, acting on production is ineffective, as it is built and programmed to respond to a demand, foreseen, imagined, or stimulated, certainly, but which proves to be real. A decrease in consumption automatically leads upstream actors to react.

Understanding the basic diversity of agricultures

It is thus necessary to recognize that, on a global scale, the diversity of agricultures is based on at least four macro-criteria:

·       The pedoclimatic environment of each country or large region that is approximately homogeneous;  

·       Their culture, in its broad sense, including agricultural and food culture, even gastronomy and culinary culture, as well as consumption modalities and the ability to resist an abundant offer;  

·       The agricultural potential of each country or region in relation to its population: net importer, net exporter, balance? What trend in the next 25 years? 

·       Their level of technical development, what evolution?

Whatever agricultural technicality is implemented, each country or region will have to take these four parameters into account, which define both levels of potential impossibility and the opening of possibilities.

The five qualitative poles of agricultural technicality development

Taking into account the diversity described above, another diversity will appear based on the understanding of the 5 qualitative poles of agricultural development, all of which are always present in all current agricultures, but not intensively similar according to regions. It is politics, here, that guides these five poles, most often without being aware of it... These five qualitative poles concerning agricultural technicality are closely intertwined. It seems erroneous to believe that there is one more important than the others. In other words, betting essentially on one of the poles, as a consequence of the current compartmentalization of sciences, or a specific ideology, will lead to failure. The question becomes: how to maintain the level of each pole according to real or supposed needs?

1)    The pole of development of a new agroscience/agricultural science (often referred to as "Agroecology") that takes into account the question of integrating agriculture into the terrestrial environment. Let us note the inherent tension in this approach; ecology is the science of relationships in a complex environment comprising a great multiplicity of species populations, while agriculture is a production system based on living organisms, or a part of the living, but it is difficult, if not impossible, to establish a boundary between "useful" living and "useless" living. We do not yet know how to conceptually combine the act of producing and the act of describing a complex system. Talking about ecosystem services is mere talk until we get into the details. Who implements them? How are they measured? How are they financed/paid? If the modalities of measurement or evaluation and the modalities of financing are too complex and bureaucratically directed, they will be ineffective.

We can refer here to the FAO slogan: producing more with less. This motto is, in fact, very complex. Moreover, the FAO elaborates on it, and "more" does not mean that there are no limits. It implies taking into account the four criteria mentioned above. Because "Less" means less area, fewer inputs, fewer negative impacts on the terrestrial environment, as well as less arduous work, etc. It is not about industrial optimization, but about an activity in alliance with the living that takes into account the factual scarcity of resources. This leads to multiple agricultures. Because who pays farmers to engage in this direction, according to what criteria? And why would they be willing to engage in such a vague direction, without guaranteed income?

How can research fit into this issue without being prescriptive or normative? How can it be supportive, accompanying, without having indicators? What indicators should be constructed? We now see agricultural research questioning its own epistemological positioning and undertaking work with farmers. Can the triptych of academic research, applied research, and development be effective without research directly in situ? What roles do the agronomy researcher, the agronomist engineer, and the farmer play?

The pole of this agro (agri)science cannot be defined without the active participation of farmers. Whatever happens, it is they who will do "the job" and they face the uncertainty of their production conditions, whether concerning the pedoclimatic environment, but also the sociotechnical and economic conditions of their production. It is best to involve them from the outset, which does not seem very visible. Reflection on the springs of action in this context is essential.

2)    We are thus led to define the second pole that I call green care, or in French "taking care of the living with which and by which agriculture produces". Here we refer directly to the recent collective work "Taking Care of Environments – Technological Design Manual, ed. M. Triclot, 2024". Agriculture produces goods, all derived from the living. The farmer must take care of the environment of his agricultural operation, but also of the environment in which his operation is inserted, knowing that the scale of this environment is still difficult to evaluate (from step to step, it can lead to the planetary scale). He must take care of his soil, the plants he cultivates, and the animals he raises, his working tools, and his clients. The "green care" is the care that every farmer takes who understands that he produces from the living, by the living, with the living. He must be very attentive to it and at the same time active. But how is he helped, financed, in this direction?

"The 'green care' is the care that every farmer takes who understands that he produces from the living, by the living, with the living."

3)    This naturally leads to defining breeding, an Anglo-Saxon term (mis)translatable into French as "improvement" of plants and animals, which are both co-producers and products. This improvement must take into account the environment in which cultivated plants and raised animals are inserted. We can define an improvement that becomes a deadlock, by "hypertely", which means, in biology, "exaggerated development of a morphological character, of an anatomical structure, which can lead to an evolutionary deadlock". Indeed, this improvement must take into account the environment: "more" does not necessarily mean "to the maximum" or "to the optimum". We know the example of the Blue White Belgian cow breed, now unable to reproduce without human intervention. Is this what is desired? For plants, this may appear different. Indeed, non-dehiscence was selected from the outset for basic reasons: it is easier to harvest from the plant than from the ground... Thus, a domesticated plant quickly loses its reproductive autonomy. What does it mean to take care of it, if not to understand its needs, given the selection pressure composed by humans? How to go further in practice? Presenting the new biotechnologies that have emerged over the past 40 years, according to this questioning, should allow us to move beyond relatively sterile ideological debates.

But it is necessary to think about policies that take into account the two previous poles. How to break out of the feedback loop that generates more "technologically orphaned" species than technologically favored species? (Cf. Megatrends in agriculture, 2018, V. Pétiard and M. Dubois). The myth of "plant breeding" is to believe that one can define an excellent variety in itself; it is up to agro(agri)sciences to express its remarkable potential! As if this potential could be defined outside of production conditions that are now transforming. The "Plant breeding" thus depends as much on the two previous poles as they depend on it.

4)    We then understand the necessity of a pole that I call precision agriculture, referring to an already well-used expression, while broadening its meaning. Precision relates to measurement: the amount of water needed where and when it is needed; the necessary action against diseases, pests, and harmful organisms, with the most precise, least costly means, just when needed; applying the necessary fertilizers just when and how they are needed. But that is not all; precision must also concern the criteria for evaluating results. Indicators, evaluation methods must meet precision criteria. This precision is not an absolute and depends on the agriculture practiced, which itself must respond: precisely what agriculture in such a place, what evaluation tool? How can evaluation be precise? How to measure environmental impact? How to precisely cross all data? How to define objectives precisely? How to measure the costs associated with implementing multi-criteria evaluations? How to describe precisely, according to adapted criteria, agricultural activity in its contextual reality? What sufficiently precise references? How to globally integrate according to precise objectives, the precisely evaluable economic performance? This requires new tools integrating (aggregating) numerous criteria/indicators, both technical, ecological, energy-related, or economic, that are adapted to each agricultural model, according to different scales and immediately understandable by farmers. This is therefore no small task!

5)    Having listed these four poles, it appears that one is missing, which I had coded under the well-known acronym NTIC-NBIC, but which I broaden to "tool-based technicality", all tool-based technicality, from the spade to the robot, from the notebook to the GPS, from the drawn description to image analysis, etc. It is a bit like a simplified dictionary, each pole refers, in fact, to the other four. But each pole also requires its own dictionary. It can be noted that this pole, like that of breeding, must avoid hypertely, which can undermine the entire productive system that is agricultural activity. For example, the construction of tools for multi-criteria analyses is now based on the acquisition and processing of an enormous amount of data. It is necessary to specify the methods and uses, in order to have coherent systems. This concerns both LCA and environmental indicators. In fact, tools are possible and are being developed, but they are energy-consuming, according to mathematical functions that are not simple and require access to a huge amount of data. At what level does the development of these tools become hypertelic to the point of undermining their development? Why use AI tools if they consume more energy than they might potentially save? How can they provide "Breeding" with a surplus of information in varietal orientation choices?

These five poles do not eliminate energy issues. Agriculture, from the beginning, produces more energy than the human who invests in it consumes. A net energy-consuming agriculture is inconceivable. Its ultimate goal is indeed to produce using solar energy. Perhaps it will be necessary to add an energy pole, a sixth, because it indeed irrigates the other five. This needs to be conceived because the entire agricultural area has a considerable energy production potential using all available techniques, that is to say, in interface with the other five poles. It was believed, more than 40 years ago, that photosynthesis and symbiotic nitrogen reduction were potentially limiting factors at the plant level; experience has shown that these two functions must be analyzed in the productive context, and sometimes the sowing density and harvest conditions are the two limiting parameters, even before the productivity of plants: a legume is not a cereal.

Given the yields currently achieved, highly optimized, wanting globally increased yields that integrate externalities will lead to new research, modeling factors not previously taken into account. Perhaps before wanting to produce more, we must first produce better. For example, what does it mean to produce more while consuming more water, which has become a scarce resource?

In conclusion, the stakes and challenges of contemporary agricultures are considerable and differ according to the policies of each region of the world that has a global agricultural policy. The complexity of agricultural production and the markets it serves implies an aggregation at the political level, in line with this complexity.

[1] According to the title of Joseph Klatzmann's (1975) work, which ended with a question mark.