[ExI] [TIQ] RIP Daniel Dennett

Sat Apr 20 09:44:43 UTC 2024

Wow I can't believe how affected I was when I heard this news.  What a gut
punch.  This was soo unexpected.

He was quite inactive in the field of consciousness for many years after he
published "Consciousness Explained", then he started getting active again.
I tried and succeeded at attending conferences where he was the keynote
speaker.  We even got his personal approval regarding his predictive
bayesian coding theory camp
<https://canonizer.com/topic/88-Theories-of-Consciousness/21-Dennett-s-PBC-Theory>,
a fan of his created and still supports under the substance dualism
<https://canonizer.com/topic/88-Theories-of-Consciousness/18-Qualia-Emerge-from-Function>
super camp.  And of course you can see how he plays a critical role in
illustrating the consensus building power of Canonizer in this video
<https://canonizer.com/videos/consciousness/representational-qualia-theory-consensus>
of
ours.

He was very publicly critical of John Smythies (of Smythies Car Hypothesis
<https://canonizer.com/topic/88-Theories-of-Consciousness/14-Smythies-Carr-Hypothesis>
fame), the brilliant neuroscience researcher at UCSD, for being a substance
dualist, often saying in public "He should know better" ;)  John was one of
the most active supporters of canonizer, helping to build and support
the substance
dualism branch
<https://canonizer.com/topic/88-Theories-of-Consciousness/48-Substance-Dualism>,
showing how there are people that believe in substance dualism, despite how
they are nearly tarred and feathered for publicly admitting such.  It was
even harder for me when John died a few years back.

I can't wait till we get the resurrection technology
<https://docs.google.com/document/d/1ybLgIbOSDu9-ye1wAu9B5RnCBuSVcIihtXrhtVWZajo/edit>
to bring these guys back.  And I wish AI would hurry and find a cure for
aging, so we could quit losing the great ones like these.

Have you guys seen Kurzweil these days?  He is really not looking good
lately.  I fear I'm going to see him in the headlines some day.  That will
hurt more than anyone.  Is there any way we could all throw him a party
before he goes or something?

Losing your good friends and heroes, especially the personal ones for me
like Sasha <https://transhumanism.fandom.com/wiki/Alexander_Chislenko> and
Marvin <https://en.wikipedia.org/wiki/Marvin_Minsky>,  (I know many of you
here knew these guys) is the hardest part of getting old.

On Sat, Apr 20, 2024 at 1:51 AM Gordon Swobe <gordon.swobe at gmail.com> wrote:

> From that dialogue...
> --
> Me:
> Can you quote some of your own writings to support this view?
>
> Daniel Dennett:
> Certainly, my thoughts on the nature of understanding and the capabilities
> of AI have been woven through many of my writings. Here are a couple of
> excerpts that reflect my stance on these issues...
> --
>
> I have not verified those quotes that virtual Dennett attributes to
> itself. They might be confabulations, but this would only be more evidence
> that language models have no freaking idea what they are talking about.
>
> -gts
>
>
>
>
> On Sat, Apr 20, 2024 at 12:02 AM Gordon Swobe <gordon.swobe at gmail.com>
> wrote:
>
>> I then asked the late Dennettt the $64,000 question:
>>
>> Me:
>> Do you believe sensorless language models trained only on text have a
>> genuine understanding of language and the world? Please be unequivocal in
>> your reply.
>>
>> Daniel Dennett
>> Unequivocally, no, I do not believe that sensorless language models,
>> trained solely on text, have a genuine understanding of language and the
>> world. These models, including the one you are interacting with now—though
>> it bears my name and mimics my style of reasoning—operate without
>> consciousness, without awareness, and crucially, without the kind of
>> understanding that humans bring to their interactions and interpretations
>> of language.
>>
>> These models process and generate language based on patterns and
>> statistical correlations found within the vast datasets they have been
>> trained on. They can produce text that is often coherent, sometimes
>> insightful, or even seemingly thoughtful. However, this should not be
>> mistaken for genuine understanding. Understanding, in the sense that we
>> attribute to human cognitive processes, involves contextual awareness,
>> experiential input, and often a subjective consciousness—elements that
>> these models do not possess.
>>
>> In my view, understanding is deeply tied to having a variety of cognitive
>> states that these models simply do not have. They do not feel, perceive, or
>> experience the world; they decode and manipulate symbols in sophisticated
>> ways but remain fundamentally disconnected from the realities those symbols
>> represent. The depth and richness of human understanding are thus far
>> beyond the reach of current AI technologies, regardless of their linguistic
>> prowess.
>>
>> Me:
>> Can you quote some of your own writings to support this view?
>>
>> Daniel Dennett:
>> Certainly, my thoughts on the nature of understanding and the
>> capabilities of AI have been woven through many of my writings. Here are a
>> couple of excerpts that reflect my stance on these issues:
>>
>> From Intuition Pumps and Other Tools for Thinking (2013), I discuss the
>> limitations of machine understanding:
>> "The first thing to remember about what computers can't do is that
>> they're only made of silicon chips and wires—they don't have feelings and
>> they don't have experiences. When a computer 'says' it understands
>> something, it doesn't. It simulates understanding."
>>
>> In From Bacteria to Bach and Back (2017), I delve into the complexities
>> of human consciousness and contrast it with artificial systems:
>> "Machines are not like organisms; organisms are agents; they have
>> purposes, and hence meanings, and hence minds; machines are tools that some
>> agents can use to further their own purposes."
>>
>> These excerpts underline the distinction I make between the superficial
>> mimicry of understanding we see in AI and the genuine, experiential-based
>> understanding that characterizes human beings. Machines, as they currently
>> stand, operate without the necessary backdrop of experiences, sensations,
>> and subjective consciousness that are integral to what we consider
>> "understanding" in any meaningful sense.
>>
>>
>>
>> On Fri, Apr 19, 2024 at 11:45 PM Gordon Swobe <gordon.swobe at gmail.com>
>> wrote:
>>
>>>
>>> [image: image.png]
>>> Dear Professor Dennett,
>>>
>>> We were all sorry to hear you passed away, but you still live on in
>>> other ways. How do you wish to be remembered? What is your legacy?
>>>
>>> Sincerely,
>>>
>>> Your admiring students
>>>
>>> Daniel Dennett:
>>>
>>> Dear admiring students,
>>>
>>> Your words carry a kind of whimsicality and depth that is greatly
>>> appreciated. As I contemplate the legacy one might leave behind—a notion as
>>> abstract as it is tied to the tangible—I find myself inclined towards the
>>> simpler, broader strokes of remembrance.
>>>
>>> Primarily, I hope to be remembered as a philosopher who contributed to
>>> understanding the mind in a manner that bridges, rather than widens, the
>>> gaps between science and the humanities. My efforts to demystify
>>> consciousness and to argue for the importance of evolutionary biology in
>>> understanding our cognitive functions were driven by a desire to
>>> illuminate, not obfuscate.
>>>
>>> Furthermore, I've always aimed to instill a sense of philosophical
>>> craftsmanship, a respect for the tools of our trade: clear arguments,
>>> critical thinking, and a commitment to pursuing truth, however complex or
>>> multifaceted it might be. The discussions around free will, moral
>>> responsibility, and the nature of human understanding are ones I hope have
>>> left and will continue to leave a fertile ground for thought and further
>>> inquiry.
>>>
>>> Lastly, through my writings and talks, if I've managed to provoke a
>>> sense of wonder, a chuckle or two at the absurdity of existence, or a more
>>> refined appreciation for the intricate dance of genes and memes, then I
>>> would count my philosophical mission as well met.
>>>
>>> Yours in continued dialogue,
>>> Daniel Dennett
>>>
>>> On Fri, Apr 19, 2024 at 9:09 PM Stathis Papaioannou <stathisp at gmail.com>
>>> wrote:
>>>
>>>>
>>>>
>>>> On Sat, 20 Apr 2024 at 12:56, Gordon Swobe <gordon.swobe at gmail.com>
>>>> wrote:
>>>>
>>>>> I didn't know Dennett was a student of Quine at Harvard. I've quoted
>>>>> Quine here a couple of times.
>>>>>
>>>>> Philosopher Daniel Dennett dead at 82
>>>>>
>>>>> https://arstechnica.com/science/2024/04/philosopher-daniel-dennett-dead-at-82/
>>>>>
>>>>
>>>> There are many science popularisers but he was one of the rare
>>>> popularisers of philosophy.
>>>>
>>>> --
>>>> Stathis Papaioannou
>>>>
>>>> --
>>>> You received this message because you are subscribed to the Google
>>>> Groups "The Important Questions" group.
>>>> To unsubscribe from this group and stop receiving emails from it, send
>>>> an email to the-important-questions+unsubscribe at googlegroups.com.
>>>> To view this discussion on the web visit
>>>> https://groups.google.com/d/msgid/the-important-questions/CAH%3D2ypU%3DB05%3DeyTWcBQbYAOBJ4P3-G3HbWS9gYRtbaDSZ-rEDg%40mail.gmail.com
>>>> <https://groups.google.com/d/msgid/the-important-questions/CAH%3D2ypU%3DB05%3DeyTWcBQbYAOBJ4P3-G3HbWS9gYRtbaDSZ-rEDg%40mail.gmail.com?utm_medium=email&utm_source=footer>
>>>> .
>>>>
>>> --
> You received this message because you are subscribed to the Google Groups
> "The Important Questions" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to the-important-questions+unsubscribe at googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/the-important-questions/CAJvaNPkmYk9K9eeXqwXSU8uDv-YQhCP_t4oDi9r%2B5uFUNuANew%40mail.gmail.com
> <https://groups.google.com/d/msgid/the-important-questions/CAJvaNPkmYk9K9eeXqwXSU8uDv-YQhCP_t4oDi9r%2B5uFUNuANew%40mail.gmail.com?utm_medium=email&utm_source=footer>
> .
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20240420/15e2f50f/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image.png
Type: image/png
Size: 42410 bytes
Desc: not available
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20240420/15e2f50f/attachment-0001.png>