In 1976, I met John Bell several times in CERN and we talked about a possible violation of optical theorem, purity tests, EPR paradox, Bell’s inequalities and their violation. In this review, I resume our discussions, and explain how they were related to my earlier research. I also reproduce handwritten notes, which I gave to Bell during our first meeting and a handwritten letter he sent to me in 1982. We have never met again, but I have continued to discuss BI-CHSH inequalities and their violation in several papers. The research stimulated by Bell’s papers and experiments performed to check his inequalities led to several important applications of quantum entanglement in quantum information and quantum technologies. Unfortunately, it led also to extraordinary metaphysical claims and speculations which in our opinion John Bell would not endorse today. BI-CHSH inequalities are violated in physics and in cognitive science, but it neither proved the completeness of quantum mechanics nor its nonlocality. Quantum computing advantage is not due to some magical instantaneous influences between distant physical systems. Therefore one has to be cautious in drawing-far-reaching philosophical conclusions from Bell’s inequalities. The true resource for quantum computing is contextuality and not nonlocality.

1976 年，我在欧洲核子研究组织 (CERN) 多次见到约翰·贝尔 (John Bell)，我们讨论了可能违反光学定理、纯度测试、EPR 悖论、贝尔不等式及其违反情况。在这篇评论中，我继续我们的讨论，并解释它们与我早期的研究有何关系。我还复制了我们第一次见面时我给贝尔的手写笔记，以及他在 1982 年寄给我的一封手写信。我们再也没有见过面，但我在几篇论文中继续讨论 BI-CHSH 不等式及其违反情况。贝尔的论文和为检验他的不等式而进行的实验所激发的研究导致了量子纠缠在量子信息和量子技术中的几个重要应用。不幸的是，它也导致了非凡的形而上学主张和推测，我们认为约翰·贝尔今天不会认可这些主张和推测。 BI-CHSH 不等式在物理学和认知科学中都被违反，但它既没有证明量子力学的完备性，也没有证明量子力学的非定域性。量子计算的优势并不是由于遥远的物理系统之间的某些神奇的瞬时影响。因此，从贝尔不等式中得出影响深远的哲学结论时必须谨慎。量子计算的真正资源是上下文而不是非局域性。