What it means to be carbon in a silicon world

‘To be yourself in a world that is constantly trying to make you something else is the greatest accomplishment,’ said Ralph Waldo Emerson (Picture source: Freepik).

Our species Homo sapiens, derives its name from the Latin word sapere, meaning ‘wisdom’. Thus we often take pride in us humans being the ‘wise man’. But what does it really mean to be a human? Does our body make us human? Well, not exactly — the ancient Egyptians and Indians routinely undertook part replacements in the form of prostheses — for medical or aesthetic reasons [1]. The earliest written evidence is said to be in the Rig Veda, an ancient, sacred poem of India written between 3500 and 1800 BC. This poem tells the story of the Warrior-Queen Vishpla who lost her leg in battle and was fitted with a prosthesis fabricated in iron to enable her to return to battle [2]. Replacing body parts does not make us any less human. Then, does our heart make us human? Not really either. Artificial hearts and pacemakers are now commonplace and have been around since Jarvik-7 in 1982 [3]. And of course we consider those with pacemakers as fellow humans. Is it the brain then, that makes us human? That is a tough one. Do we think and feel in our brains? Is that our identity as a species? Philosophers, scientists, and poets alike, have tried to address this conundrum since time immemorial. In 500 BCE, Parmenides, the famous Greek philosopher stated, ‘Thought and being are the same,’ thereby linking thinking and existence [4]. Plato and Aristotle took this a notch higher by maintaining in 350 BCE, ‘we are aware of our thinking, hence are conscious.’ To the layer of existence, they thus compounded the definition with attaching to it, that of experience and consciousness [5]. Adi Shankara, the outstanding Indian philosopher saint further stated in 700 CE, ‘one’s existence cannot be doubted, for no one thinks I am not’, thus adding another profound layer of self-reflection to the definition of being a human [6]. And to bring us back to where it all began, Rene Descartes closed the loop back almost 2000 years by stating in 1700 CE, one of his most iconic statements [7], ‘je pense, donc je suis’ or ‘I think therefore I am.’ Thinking, experiencing, consciousness and self-reflection seem to have been the recurring theme in our own definition of what makes us us.

Philosophical and metaphysical discussions notwithstanding, for a long time, common scientific wisdom had been that three characteristic traits make humans unique — bipedalism, tool making, and language [8]. Let us address each of these in the context of some of the latest developments in technology.

1. Bipedalism: The last decade has seen an explosion of companies in the field of robotics [9]. But despite the exuberance, for the overall sector, it has been a reasonably mixed bag [10]. While the market may be facing a slowdown, the technological progress made in robots with human-like capabilities has been astounding to say the least. Bipedal robots are learning to move with arms as well as legs to get around tricky environments [11]. And though, nothing to beat Usain Bolt, Cassie, a bipedal robot has managed to clock a stable speed of 9 mph [12]. As stated by the lead researcher of the study, ‘This may be the first bipedal robot to learn to run, but it won’t be the last.’ It is not just speed, but also dexterity that is improving. Boston Dynamics’ Atlas, keeps getting eerily better at tasks and working alongside humans [13]. So, machines are starting to work towards leveling up with us and it is conceivable that the future will see significant progress. Probably, we won’t be unique in bipedalism.
2. Tool making: Originally thought to be a skill possessed only by humans, tool making requires a sophisticated level of cognition. But recent research has shown that apart from primates, other species that use tools include corvids, elephants, dolphins, otters, and octopuses (or octopodes, if you will) [14]. This is not limited only to fellow living occupants of the planet. Now, with significant advances in embedded systems and machine learning, robots have started to be able to assemble tools for specific needs, albeit simple ones [15]. A couple of years ago, one of the earlier demonstrations named RoboGyver, after the 1980s action-adventure TV series about Angus MacGyver, had already promised an interesting start to synthetic tool-making skills. And of course, the skills have been advancing by the year [16]. Given time, it seems likely we’ll find robots as useful lab assistants in the future — conceiving and conducting the experiments for us. So, we’re closing the gap here too. Therefore, tool-making skills may not qualify as a differentiator in being classified as a human either.
3. Language: Down to the last differentiator, let us see how strong this one holds. At the outset, one needs to distinguish between communication and language. To the purists, animals do certainly communicate, but they don’t possess language skills in the way humans do. They allow their bearers to express only a small subset of the concepts they can remember, represent and manipulate productively [17]. While language is certainly not logic, the two are related more closely than would seem apparent to most of us — they may be considered fruits of the same tree. It is said that in order to master basic logic, one must understand negation. The inability of animals to use negation shows they lack basic logical abilities. Another essential characteristic of our language is its normativity — namely, the fact that there are right and wrong uses of a word or phrase. As a recent study states [18], ‘Animals produce sounds that express their emotions, and some can use signs in a Pavlovian way, as a result of an association between previous uses and succeeding events. But without “Why?” and “No!” there’s nothing resembling human language.’ What about machines? Several studies made publicly available show the surprising ability of computer programs to start forming rudimentary forms of language when communicating with each other and working towards a common objective [19, 20]. I use the word rudimentary on purpose, for, while they have not really demonstrated language skills in the strictest definition, their ability to effectively engage with each other may be more akin to the communication of animals, talked about previously. As language models and machine learning become more sophisticated, there is a clearer emergence of a form of intra- and inter-program engagement taking place within the process steps, that needs us humans to interpret and explain the flow. The field of explainable AI (XAI) [21], while in its nascency, explores this very translation, of the inner happenings of a code developed in a manner that humans can understand — why did the code churn out this answer? What were its unwritten considerations under which it is working? Why did the code choose to optimize a certain parameter first, over another? But then again, correlations and causations are not the same. Does it suffice for a system to develop sophisticated abilities of discovering patterns of correlations or does it need to understand the root cause? XAI may talk about the root cause, but can the machines themselves posit one when posed the question? ‘Most AI systems capture correlations and associations in data and not causal relationships. Explanations of the AI systems’ predictions would make the correlations more transparent. They would not, however, make the explained relationships causal [22].’ Causal understanding requires an appreciation of what is called context [23]. It is widely accepted that language requires context in order to function as communication between speakers and listeners — something which has been out of grasp of such machines — context based learning [24].’ Fast forward the clock to 2023. With the advent of a combination of generative pre-trained transformers [25, 26] and advances in context-based learning [27], an example of which is ChatGPT [28], the race towards the new frontier of AI — Language — has been unlocked [29]. So, there it goes! Maybe it is conceivable that in the future, the gap between humans and machines, in their ability to develop and use a language starts to close. So, we’re now really left struggling to figure out what makes us human!

Keeping in mind that a lot of this progress of synthetic systems being able to catch up to humans has been brought about by the invention of the silicon transistor, we must ask ourselves the question of what makes us carbon beings unique in this silicon world? It is without a doubt, that the world will become increasingly more digitally connected and reliant on the wonders of silicon. But we’re carbon, not silicon. There must be something collective about us, which is greater than each of us and that makes us so successful as a species.

In an article published in 2018 in the journal Scientific American, an interesting point was brought up, by the author [30]. ‘The emerging consensus is that humanity’s accomplishments derive from an ability to acquire knowledge and skills from other people. Individuals then build iteratively on that reservoir of pooled knowledge over long periods. This communal store of experience enables creation of ever more efficient and diverse solutions to life’s challenges. It was not our large brains, intelligence or language that gave us culture but rather our culture that gave us large brains, intelligence and language. For our species and perhaps a small number of other species, too, culture transformed the evolutionary process.’ The article goes on to state, ‘Why, then, don’t other primates have complex culture like us? Why haven’t chimpanzees sequenced genomes or built space rockets? Mathematical theory has provided some answers. The secret comes down to the fidelity of information transmission from one member of a species to another, the accuracy with which learned information passes between transmitter and receiver. The size of a species’ cultural repertoire and how long cultural traits persist in a population both increase exponentially with transmission fidelity. Above a certain threshold, culture begins to ratchet up in complexity and diversity. Without accurate transmission, cumulative culture is impossible. But once a given threshold is surpassed, even modest amounts of novel invention and refinement lead rapidly to massive cultural change. Humans are the only living species to have passed this threshold.’ Humans thus have this culture, that at the present moment, machines seem to be lacking by far.

Thus, it all boils down to information transmission from one member of a species to another in a manner that builds upon the passed down knowledge, such that the sum becomes greater than its parts. Maintaining and building up this human culture via effective human-human interaction is absolutely essential to retain the sense of being a human. In fact, it has been recently pointed out that culture helps humans adapt to their environment and overcome challenges better and faster than genetics, and that the importance of culture surpasses the value of genes as the primary driver of human evolution [31]. Having said that, the news does not seem to be that good — cultural diversity is disappearing quickly [32]— the study reveals. We all know what happens when genetic diversity decreases — the species becomes very prone to being wiped out by a dedicated pathogen. Now imagine what happens when cultural diversity decreases or is lost — do we lose our humanness?

By maintaining healthy relationships that enables the authentic and faithful transmission of information between us and other humans, especially as an input from our previous generation, modified with our own learnings and transmitted to our future generations, we retain what is human in us. And as long as we can keep this vast diversity of cultural heritage across our species alive, there is a clear place for us. For we then become a link in the long chain of human history. We then continue to carry forward in us, a portion of our unadulterated humanness connecting us to the origin of our species. As Lyndon Johnson stated when signing the Wilderness Act in 1964, ‘If future generations are to remember us with gratitude rather than contempt, we must leave them something more than the miracles of technology. We must leave them a glimpse of the world as it was in the beginning, not just after we got through it [33].’

So, then maybe after all, this is what it means to be carbon in an increasingly silicon world — being an active participant in the cumulative cultural heritage that lies at the heart of what eventually makes us human.

(Disclaimer: all views expressed in this are my own and do not represent those of any organization that I am affiliated with)

Biblography:

1. Pine, Keith R.; Sloan, Brian H.; Jacobs, Robert J. (2015). Clinical Ocular Prosthetics. Springer. ISBN 9783319190570.
2. Thurston AJ. Paré and prosthetics: the early history of artificial limbs. ANZ J Surg. 2007 Dec; 77(12):1114–9. doi: 10.1111/j.1445–2197.2007.04330.x. PMID: 17973673.
3. https://www.si.edu/object/jarvik-7-artificial-heart%3Anmah_1147093
4. https://plato.stanford.edu/entries/parmenides/
5. https://www.britannica.com/biography/Aristotle/Philosophy-of-mind
6. https://digitalassets.lib.berkeley.edu/main/b20782842_C020830084.pdf
7. https://www.historyofinformation.com/detail.php?id=1923
8. https://education.nationalgeographic.org/resource/what-makes-us-human/
9. https://www.mckinsey.com/industries/industrials-and-electronics/our-insights/unlocking-the-industrial-potential-of-robotics-and-automation
10. https://www.forbes.com/sites/forbesbusinesscouncil/2021/07/09/the-demise-of-robotics-companies-learning-from-past-mistakes
11. https://spectrum.ieee.org/bipedal-robot-learning-to-move-arms-legs
12. https://www.theverge.com/2022/9/30/23380061/bipedal-robot-100-meter-record-run-time-cassie-agility-robotics-osu
13. https://arstechnica.com/gadgets/2023/01/boston-dynamics-atlas-robot-grows-a-set-of-hands-attempts-construction-work/
14. https://www.popularmechanics.com/science/animals/g39714258/animals-using-tools/
15. https://www.asme.org/topics-resources/content/robot-rises-to-the-challenge-of-toolmaking
16. https://www.technologyreview.com/2019/04/11/65893/a-robot-has-figured-out-how-to-use-tools/
17. https://royalsocietypublishing.org/doi/10.1098/rstb.2019.0046
18. https://blogs.scientificamerican.com/guest-blog/can-animals-acquire-language/
19. https://www.theatlantic.com/technology/archive/2017/06/artificial-intelligence-develops-its-own-non-human-language/530436/
20. https://nypost.com/2022/06/03/artificial-intelligence-spotted-inventing-its-own-creepy-language/
21. https://www.darpa.mil/program/explainable-artificial-intelligence
22. https://www.sciencedirect.com/science/article/pii/S2666389922002677
23. Cohen, A.O., Nussenbaum, K., Dorfman, H.M. et al. The rational use of causal inference to guide reinforcement learning strengthens with age. npj Sci. Learn. 5, 16 (2020). https://doi.org/10.1038/s41539-020-00075-3
24. https://pubmed.ncbi.nlm.nih.gov/30784019/
25. https://blogs.nvidia.com/blog/2022/03/25/what-is-a-transformer-model/
26. https://www.mckinsey.com/featured-insights/mckinsey-explainers/what-is-generative-ai
27. https://towardsdatascience.com/advancing-machine-intelligence-why-context-is-everything-4bde90fb2d79
28. https://openai.com/blog/chatgp
29. https://www.forbes.com/sites/robtoews/2022/02/13/language-is-the-next-great-frontier-in-ai/
30. https://www.scientificamerican.com/article/what-made-us-unique/
31. Timothy M. Waring, Zachary T. Wood. Long-term gene–culture coevolution and the human evolutionary transition. Proceedings of the Royal Society B: Biological Sciences, 2021; 288 (1952): 20210538 DOI: 10.1098/rspb.2021.0538
32. Zhang, H., & Mace, R. (2021). Cultural extinction in evolutionary perspective. Evolutionary Human Sciences, 3, E30. doi:10.1017/ehs.2021.25
33. https://www.nps.gov/romo/wilderness_act_turns_40.htm