Deep Ecology and AI
Including Ecology When Thinking Technology
The previous few days I have written about the direct consequences of expanding technological infrastructure, as it is not a practice without potential issues. Especially due to the need for data centres both when it comes to electricity and water. The great increase of data centres, and machine learning applications that proposes to greatly increase the collection of large datasets does of course have an influence on the environment.
Human is part of nature, one of several beings on the planet.
It is perhaps ironic that within the field of artificial intelligence we talk of deep learning, as one technique, when we often fail to learn about a broader ecology. One may come to realise to some extent it seems more like shallow learning when we fail to take these considerations.
Deep learning is part of a broader family of machine learning methods based on artificial neural networks with representation learning.
Let us contrast or complement this deep learning with deep ecology.
Deep ecology is an ecological and environmental philosophy promoting the inherent worth of living beings regardless of their instrumental utility to human needs, plus a restructuring of modern human societies in accordance with such ideas.
We have to consider our thoughts about nature. Remembering what lies beyond the immediate representations. Many graphs in consumption and emission do still point upwards, yet there is a world beyond graphs in our habits.
Already in the 90s, Arne Næss was questioning whether we should stop flying as much as we do, or ask ourselves deeper questions in terms of our relationship with nature.
I first encountered the name Arne Næss when I went to the Arne Næss lectures in Oslo. The discussions there sparked thoughts that I had not considered before. I remember asking Evgeny Morozov who was lecturing at the time a question when he held his lecture there. I asked him about AI back then, whether we should grant right to robots.
He answered back that maybe I should consider other rights first. He talked then of how the attention is diverted away from important questions regarding technology and data centres, with ownership in particular as an issue.
His Arne Næss lecture at our University of Oslo was called Resisting Data Extractivism: Towards Humane Digital Alternatives. Has the time come to be more critical of data ownership, and to ask whose representation of the future we are envisioning. Who crafts the digital narratives that in turn becomes symbols shaping our culture and actions?
With my background interest in sustianability and entrepreneurship I started thinking there might be other issues we ignore when we are caught up in future tense discussions of machine intelligence.
As such I came to hear the name Arne Næss by proxy, and entering an auditorium for ethics lectures. Arne Næss did interest me to a greater degree, so I started attempting to understand who he was.
Arne Dekke Eide Næss was a Norwegian philosopher who coined the term “deep ecology” and was an important intellectual and inspirational figure within the environmental movement of the late twentieth century.
Arne Næss (1917–2009) was the youngest person to be appointed full professor at the University of Oslo (UiO) and the only professor of philosophy in the country at the time. He was a professor at UiO in the time period 1939 til 1970. He led the UNESCOs research project about ideology1948–1949, and in 1960 he started the scientific journal Inquiry.
Yet I am interested to trace part of these thoughts before any description.
Early History of Arne Næss & Deep Ecology
Environmentalism had emerged as a popular grassroots political movement in the 1960s. Some say this was with the publication of Rachel Carson’s book Silent Spring.
“Silent Spring is an environmental science book by Rachel Carson. The book was published on September 27, 1962, documenting the adverse environmental effects caused by the indiscriminate use of pesticides. Carson accused the chemical industry of spreading disinformation, and public officials of accepting the industry’s marketing claims unquestioningly.”
Arne Næss has claimed this book was a key influence in his vision of deep ecology.
The term ecological wisdom, synonymous with ecosophy, was introduced by Næss in 1973. He argued that ecologically responsible policies are concerned only in part with pollution and resource depletion. There are deeper concerns which touch upon principles of diversity, complexity, autonomy, decentralization, symbiosis, egalitarianism, and classlessness.
“In his essay The Shallow and the Deep, Long-Range Ecology Movements: A Summary, published in 1973 in the journal Inquiry, Norwegian philosopher Arne Næss (1912–2009) coined the concept deep ecology. Therein, he argued that only a “deep” transformation of modern society could prevent an ecological collapse. Næss criticized one-sided technological approaches in dealing with environmental problems, an attitude he called shallow ecology. Instead, the design of a sustainable world should be seen not only as a question of environmental technology and economy, but also as an issue of worldviews and attitudes toward life. The ideas of deep ecology soon became adopted by environmental activists and academics. Today, they play an important role in reflections on global environmental ethics.”
Environment & Society
Therein, he argued that only a “deep” transformation of modern society could prevent an ecological collapse. Næss criticized one-sided technological approaches in dealing with environmental problems, an attitude he called shallow ecology. Instead, the design of a sustainable world should be seen not only as a question of environmental technology and economy, but also as an issue of worldviews and attitudes toward life.
I think these words from Arne Næss in 1973 resonate into the present:
“Principles of diversity and of symbiosis. Diversity enhances the potentialities of survival, the chances of new modes of life, the richness of forms. And the so-called struggle of life, and survival of the fittest, should be interpreted in the sense of ability to coexist and cooperate in complex relationships, rather than ability to kill, exploit, and suppress.”
He argued too that we had to go beyond simply talking of resource depletion.
To me I find this still holds true, we tend to focus on discussing resources when we should ask ourselves deeper questions.
As Greta Thunberg said in 2019:
“I know we need a system change rather than individual change. But you can not have one without the other.”
We must think of life on the planet in broader terms.
For me it connects to moments in my life. Thinking of seeing a hedgehog walking across the street with its two children, or the taste of a river.
“By an ecosophy I mean a philosophy of ecological harmony or equilibrium. A philosophy as a kind of sofia wisdom, is openly normative, it contains both norms, rules, postulates, value priority announcements and hypotheses concerning the state of affairs in our universe. Wisdom is policy wisdom, prescription, not only scientific description and prediction.” [italics in original, from p. 99 of Næss 1973 article]
He argued too in the article for a global approach with regional differences.
The ideas of deep ecology soon became adopted by environmental activists and academics. Today, they play an important role in reflections on global environmental ethics.
He thereby distinguished between what he called deep and shallow ecological thinking. In contrast to the prevailing utilitarian pragmatism of western businesses and governments, he advocated that a true understanding of nature would give rise to a point of view that appreciates the value of biological diversity, understanding that each living thing is dependent on the existence of other creatures in the complex web of interrelationships that is the natural world.
So for which reasons is this relevant to the development of AI?
Purposeful AI for Our Shared Ecology
Deep ecology is normative according to Næss in 1973, and think each AI application should be asked normative questions in terms of its purpose. If we are to build solutions they must be responsible not only from a resources or depletion perspective. In my clear opinion AI applications must contribute to restoration of natural habitats. If we think technology applications are neutral we are making a grave mistakes that has been perpetuated for too long. There is no implementation of technology that I have come across that is neutral, so we should clear about the purpose for which we apply technology or not.
To me it has to be an overall mindset, however I think there are some larger questions that developers or people involved in AI can ask when approaching an application of machine learning techniques. I have added one more from the previous list taking the question of ecology in mind.
- Considering the question of ecology what is a good application of AI?
- Do I need applied AI here?
- Could I solve this problem in any other way?
- If I make this solution does it need almost constant cloud connectivity to run and how much will this influence emissions?
- How efficient do I need my solution to be and what is the tradeoff in terms of environmental costs in release of carbon or damage to the environment?
- Where is my datacenter, and how much data am I storing? Do I need all this data or am I just storing because I do not know what to use.
- How can I help other developers, companies, governments and nonprofits reduce their use of machine learning techniques in an area where it may consume too much power or not be needed in the first place?
Of course there is to some extent a resources perspective in this line of questioning, and it could be improved. The first question might be a bit mysterious referring to a question of ecology that is not expressed or explained, so I would be open for thoughts on this matter too.
This is #500daysofAI and you are reading article 272. I am writing one new article about or related to artificial intelligence every day for 500 days. My current focus for 100 days 200–300 is national and international strategies for artificial intelligence.
