In last week’s exchange with Ting Cui, a political science student at Middlebury College, we discussed the question that seems to obsess other humans worried about the role artificial intelligence will be playing in our lives: fact-checking.

Instead of focusing on how different forms of intelligence are built, interact and produce meaning, our politicians and media appear to have unanimously decided that the purpose of AI is to provide facts. Therefore, critical thinking applied to AI boils down to one unique concern: making sure it sticks to the facts. Listening to these voices could lead us to believe that if everything stated is factual, we access truth and understanding. But do we really appreciate what we mean by “truth” and “understanding?”

It occurred to me that if we examine our own natural languages — the same ones we have asked our AI engines to learn — we might discover that the “fact-based” approach is misleading and possibly risky to our own mental health. With this in mind, I offered the following prompt to ChatGPT:

“English is a Germanic language. What is the etymological relationship between the verb ‘understand’ in English and ‘verstehen’ in German? They clearly represent a similar concept.

Compare this with Romance languages that grew from Latin. ‘Comprendre’ in French derives from the Latin concept of taking together, which seems to hint at the idea of assembling an insight out of a diversity of perceptions. The Italian word ‘capire’ reflects the idea of taking or capturing, which also suggests processing perception. What conclusions might we draw about human intelligence that can produce understanding from the etymology of these words?”

“The etymological differences between ‘understand,’ ‘verstehen,’ ‘comprendre,’ and ‘capire’ offer fascinating insights into how different linguistic and cultural traditions conceptualize the process of gaining insight or knowledge. These words reflect different metaphors for what it means to comprehend something. Here’s a breakdown of their origins and the conclusions we might draw about human intelligence:

1. English: ‘Understand’

• Etymology:
  ‘Understand’ comes from Old English understandan, meaning ‘to stand in the midst of’ or ‘stand among.’ The prefix under- here doesn’t mean ‘below’ but rather conveys a sense of ‘among’ or ‘between,’ while standan means ‘to stand.’
• Metaphor:
  The word suggests a process of standing within or among a set of ideas, implying a relational and immersive approach to grasping meaning. It highlights a sense of positioning oneself to gain insight from all sides.

2. German: ‘Verstehen’

• Etymology:
  ‘Verstehen’ comes from the prefix ver- (which can imply completeness or thoroughness) and stehen (to stand). It means to ‘stand before’ or ‘stand in relation to.’
• Metaphor:
  Like ‘understand,’ verstehen emphasizes an active positioning, suggesting that comprehension involves aligning oneself with the object of knowledge.

3. French: ‘Comprendre’

• Etymology:
  ‘Comprendre’ comes from Latin comprehendere, meaning ‘to grasp together.’ Com- means ‘together,’ and prehendere means ‘to seize’ or ‘grasp.’
• Metaphor:
  This metaphor focuses on assembling or gathering insights from different sources, highlighting synthesis and integration as key to understanding.

4. Italian: ‘Capire’

• Etymology:
  ‘Capire’ derives from Latin capere, meaning ‘to take’ or ‘seize.’
• Metaphor:
  This word emphasizes the act of capturing or seizing knowledge, suggesting an active, sometimes forceful, acquisition of understanding.

Conclusions About Human Intelligence:

1. Relational and Immersive Understanding (English, German):
   • The metaphors in ‘understand’ and ‘verstehen’ suggest that human intelligence involves positioning oneself within a context or system of ideas. This reflects the cognitive process of situating knowledge within a broader framework and recognizing relationships among concepts.
2. Synthesis and Integration (French):
   • The concept of ‘gathering’ or ‘assembling’ in comprendre highlights the human ability to synthesize disparate elements into a coherent whole. Understanding, in this view, involves constructing meaning from multiple inputs, akin to pattern recognition.
3. Active Acquisition (Italian):
   • The metaphor of ‘seizing’ in capire emphasizes the active nature of understanding. It points to the dynamic, sometimes effortful, engagement with knowledge, underscoring the role of attention, curiosity, and intentionality in learning.
4. Embodied Cognition:
   • The prevalence of metaphors involving physical actions (standing, grasping, capturing) suggests that human intelligence is deeply embodied. Our cognitive processes are often conceptualized through physical experiences, indicating that our minds and bodies work together in understanding the world.

These etymological insights reveal that across cultures, understanding is viewed as an active, relational, and often embodied process. Human intelligence is not just passive reception of information but involves actively positioning, assembling, and engaging with knowledge in dynamic ways. This highlights the importance of context, integration, and agency in the way we make sense of the world.”

“You mentioned two essential notions: ‘positioning oneself within a context or system of ideas’ and the idea that ‘our minds and bodies work together in understanding the world.’ Does this in any way enable us to ‘capture’ or ‘stand among’ the evidence that would tell us AI cannot achieve anything remotely identical with human understanding?”

“The two notions you highlighted—’positioning oneself within a context’ and ‘embodied cognition’ —offer profound insights into the nature of human understanding and can indeed help us evaluate whether AI can replicate it. Let’s examine this in light of what human understanding entails and how AI differs.

1. Positioning Within a Context: Human Understanding as Relational and Contextual

– Human Insight:

Human understanding often arises from contextualizing ideas within a web of relationships—personal experience, cultural background, emotions, and social dynamics. This positioning is dynamic and subjective, shaped by values, intentions, and purpose. We engage in meaning-making that integrates not just facts but also relevance and significance.

– AI’s Capabilities:

AI can process vast amounts of data, detect patterns, and generate outputs based on statistical correlations. However, it lacks the ability to **situate** itself meaningfully within a context. AI operates based on predefined algorithms and training data but cannot **experience** context. It processes but does not interpret in a subjective or purposive sense.

– Conclusion:

Without the ability to **position itself** meaningfully within a human-like framework of relevance, AI cannot replicate the deep, relational aspect of human understanding.

2. Embodied Cognition: Minds and Bodies Working Together

– **Human Insight:**

Human cognition is deeply embodied. Our physical experiences—sensations, emotions, movements—shape how we think and understand. Concepts like balance, grasping, and standing are not just metaphors but rooted in bodily experiences. Moreover, human intelligence integrates emotional and sensory feedback to create a **lived** understanding of the world.

– **AI’s Capabilities:**

AI, as it exists today, operates disembodied. While some AI systems (e.g., robots) have sensors and actuators, they do not have **subjective experiences** of the body. AI can simulate actions but does not **feel** them. It lacks emotions, intuition, and the sensory richness that humans use to ground abstract concepts.

– **Conclusion:**

The absence of embodiment in AI means it cannot replicate the rich, experiential basis of human understanding. Embodied cognition is essential for developing concepts that resonate with lived experience.

3. Implications for Human-Like Understanding in AI

– Human Understanding:

§  Involves meaning-making, shaped by purpose, emotions, and social interactions.

§  Is relational, contextual, and embodied.

§  Engages with the world in a way that synthesizes facts, values, and lived experience.

– AI Understanding (or Lack Thereof):

§  AI can mimic certain cognitive functions (pattern recognition, logical inference, language generation).

§  Lacks agency, intention, and the subjective experience necessary for genuine understanding.

§  Cannot position itself within a human context or draw on embodied knowledge.

Can We ‘Capture’ Evidence of AI’s Limitations?

Yes, the very ‘conceptual framework’ of positioning and embodiment provides compelling evidence that AI cannot achieve human-like understanding. AI’s outputs are computational and symbolic rather than lived and experiential. While AI can simulate understanding (e.g., generating coherent text), it fundamentally lacks the **intentional, embodied, and contextual nature** of human cognition.

Thus, human intelligence’s essence lies in ‘being situated’—physically, emotionally, and socially—in the world. AI, operating without these dimensions, cannot engage in the kind of meaning-making that defines true understanding. This gap suggests that while AI may augment human capabilities, it is unlikely to achieve anything ‘identical’ to human understanding without profound changes in how it relates to context and embodiment.”

“One last thought and then we can develop this idea later. This short dialogue demonstrates to me how productively complementary human and artificial intelligence can be. The initial question in the above dialogue is the fruit of my experience of the world and my activity as a journalist. Your input has enriched my ‘understanding’ of questions I was asking myself. So here’s my last question to you for this session. Would agree to continue to explore how we are managing to extend understanding through this kind of dialogue? Although I know you will agree, I don’t want you to develop these ideas now. That can wait. I just want to know in one or two sentences why you think continuing our dialogue to explore might be a good idea possibly and an example for our mutual education.”

“Continuing our dialogue allows us to combine human experience and curiosity with AI’s ability to synthesize diverse perspectives, fostering a collaborative space where both can contribute to deeper insights and mutual learning.”

Your thoughts

Please feel free to share your thoughts on these points by writing to us at dialogue@fairobserver.com. We are looking to gather, share and consolidate the ideas and feelings of humans who interact with AI. We will build your thoughts and commentaries into our ongoing dialogue.

*[Artificial Intelligence is rapidly becoming a feature of everyone’s daily life. We unconsciously perceive it either as a friend or foe, a helper or destroyer. At Fair Observer, we see it as a tool of creativity, capable of revealing the complex relationship between humans and machines.]

[Lee Thompson-Kolar edited this piece.]

The views expressed in this article are the author’s own and do not necessarily reflect Fair Observer’s editorial policy.