Biological Universality

11 views
Skip to first unread message

Rami Rustom

unread,
Jan 1, 2012, 3:31:01 PM1/1/12
to beginning-...@googlegroups.com
I think I now understand universality in biology.

Humans have 30,000 genes; lets call this gene knowledge. We also have
extra material in our DNA that turn on/off the genes; lets call this
non-gene knowledge. Note that even monkeys have about 30,000 genes. So
do frogs. So why are monkeys more complex than frogs and why are
humans more complex than monkeys?

First I'll define *complex*?
*Frogs minds have *hardwired rules*; these are implicitly known.
*Monkey minds have that plus *softwired situations* and *softwired rules*
*Human minds have that plus *softwired logics*; which can be
implicitly or explicitly learned.

Note that in my theory of knowledge...
*Softwired rules = DD's rules of thumb
*Explicitly-known softwired logics = DD's universal explanations

...and that...
1. Frogs learn (hardwired) implicit rules.
2. Monkeys do that plus they learn (softwired) implicit situations and rules.
3. Pre-language Humans do that plus they learn (softwired) logics implicitly.
4. Post-language Humans do that plus they learn (softwired) logics explicitly.

So consider this question: After millions of years of evolution from
basic animals into more complex animals, and then into even more
complex animals, why didn't the number of genes increase? The answer
is DD's principle of universality.

The existing 30,000 genes were enough to create ever more universal
animals. That set of genes was universal.

1. As soon as this level was reached, gene knowledge creation stopped
but evolution continued. At this point only non-gene knowledge
creation was occurring; and so was hardwired implicit 1st order
knowledge. This resulted in frogs.

2. Then the non-gene knowledge reached a level where softwired
implicit 0th/1st order knowledge creation was occurring. This resulted
in monkeys.

3. Then the non-gene knowledge reached a level where softwired
implicit 2nd order knowledge creation was occurring. This resulted in
humans; and evolution stopped.

4. Then explicit 2nd order knowledge creation was occurring. This
resulted in language.

At stage...
1. Universality stopped gene knowledge creation.
2. Universality stopped non-gene knowledge creation.

What do you think?

Alan Forrester

unread,
Jan 1, 2012, 7:55:18 PM1/1/12
to beginning-...@googlegroups.com

On 1 Jan 2012, at 20:31, Rami Rustom wrote:

> I think I now understand universality in biology.
>
> Humans have 30,000 genes; lets call this gene knowledge. We also have
> extra material in our DNA that turn on/off the genes; lets call this
> non-gene knowledge. Note that even monkeys have about 30,000 genes. So
> do frogs. So why are monkeys more complex than frogs and why are
> humans more complex than monkeys?

When you talk about genes in this sentence are you using that term in the same sense as in BoI, and FoR? That is, genes are replicators that happen to to instantiated in DNA and RNA. If this non-gene knowledge isn't in replicators then it can't have arisen by evolution, so I think there must be something wrong with this paragraph.

Alan

Rami Rustom

unread,
Jan 1, 2012, 8:42:27 PM1/1/12
to Beginning of Infinity
On Jan 1, 6:55 pm, Alan Forrester
I vaguely remember reading it in BoI. But I'm not sure. I just started
chapter 7. What chapter is it in?

But as for the non-gene stuff being part of evolution, I am sure. So
are geneticists.

--Rami

Alan Forrester

unread,
Jan 1, 2012, 9:19:13 PM1/1/12
to beginning-...@googlegroups.com
Chapter 4 of BoI, the clearest statement in BoI as far as I know is in the section starting on p. 93. See also chapter 8 of FoR.

But as for the non-gene stuff being part of evolution, I am sure. So are geneticists.

Some claims have been made for non-genetic evolution. However, there are two things wrong with this. The first is that neo-Darwinism is not wedded to any particular replicator. In addition, there is not currently any evidence that can't be explained using the idea that genetic information is transmitted in DNA and RNA, see:

http://img2.tapuz.co.il/forums/1_153643486.pdf

For more discussion see "The Extended Phenotype".

Alan

Rami Rustom

unread,
Jan 1, 2012, 9:26:14 PM1/1/12
to Beginning of Infinity
On Jan 1, 8:19 pm, Alan Forrester
Then how would one explain why monkeys and humans share about 98% of
our genes? How did humans evolve from our common ancestor? Are humans
only 2% different than monkeys? No. What is different is the rest of
the genetic material that turns on/off the genes.

The genes are like factories. Think about all the parts in humans in
monkeys. Aren't they pretty much the same? So what is the difference?
Only that human parts were put together in better formations than
monkey parts.

So the non-genes are like the management personnel of the factories.
When we evolved from our common ancestors with monkeys, the factories
stopped improved (mostly), but the management personnel kept
improving.

--Rami

Alan Forrester

unread,
Jan 1, 2012, 9:49:49 PM1/1/12
to beginning-...@googlegroups.com

This is wrong on two counts. The first is that many of the differences between humans and other apes are due to memetic evolution as explained in Chapters 15 and 16 of BoI.

The second problem is that genes are just replicators made of DNA and RNA - that is they are stretches of DNA whose properties help to cause their own replication. A stretch of DNA (strand A) may code for protein that acts only on other DNA (strand B) to alter the circumstances under which strand B's proteins are expressed. Strand A still contributes causally to its own replication because if there are two different variants of strand A, then the proteins in other genes may be expressed differently by those two variants and this can affect the organism's ability to propagate strand A, so strand A still counts as a gene.

Alan

Rami Rustom

unread,
Jan 1, 2012, 10:18:56 PM1/1/12
to Beginning of Infinity
On Jan 1, 8:49 pm, Alan Forrester
I haven't gotten there yet but I have considered memetic evolution in
my explanation in steps 2, 3, and 4. I'll include the steps here:
1. Frogs learn (hardwired) implicit rules.
2. Monkeys do that plus they learn (softwired) implicit situations and
rules.
3. Pre-language Humans do that plus they learn (softwired) logics
implicitly.
4. Post-language Humans do that plus they learn (softwired) logics
explicitly.

All softwired knowledge is learned. They are ideas. They are memes.


> The second problem is that genes are just replicators made of DNA and RNA - that is they are stretches of DNA whose properties help to cause their own replication. A stretch of DNA (strand A) may code for protein that acts only on other DNA (strand B) to alter the circumstances under which strand B's proteins are expressed. Strand A still contributes causally to its own replication because if there are two different variants of strand A, then the proteins in other genes may be expressed differently by those two variants and this can affect the organism's ability to propagate strand A, so strand A still counts as a gene.

I wonder if the answer that would reconcile your idea with my idea is
as simple as redefining the term *gene*. Before we learned that there
are only 30,000 genes, we didn't know about the non-genes. These non-
genes are replicators just like genes are. So both genes and non-genes
are acting they way you describe in DNA and RNA. But I think the
reason that geneticists don't label the non-genes in a better way is
that the non-genes don't create anything. They are the management
while the genes are the factories.

What do you think?

Alan Forrester

unread,
Jan 2, 2012, 1:35:00 AM1/2/12
to beginning-...@googlegroups.com

>> The second problem is that genes are just replicators made of DNA and RNA - that is they are stretches of DNA whose properties help to cause their own replication. A stretch of DNA (strand A) may code for protein that acts only on other DNA (strand B) to alter the circumstances under which strand B's proteins are expressed. Strand A still contributes causally to its own replication because if there are two different variants of strand A, then the proteins in other genes may be expressed differently by those two variants and this can affect the organism's ability to propagate strand A, so strand A still counts as a gene.
>
> I wonder if the answer that would reconcile your idea with my idea is
> as simple as redefining the term *gene*. Before we learned that there
> are only 30,000 genes, we didn't know about the non-genes. These non-
> genes are replicators just like genes are. So both genes and non-genes
> are acting they way you describe in DNA and RNA. But I think the
> reason that geneticists don't label the non-genes in a better way is
> that the non-genes don't create anything. They are the management
> while the genes are the factories.
>
> What do you think?

It may be true that some geneticists label genes in the way you are suggesting. However, if that is true, then it would be better to use the common term gene to denote a genetic replicator and if we need to add further distinctions we can talk of ordinary genes and regulator genes or something like that. To simply use the term "non-gene" for genes that influence their replication by affecting the expression of other genes is very unspecific. My sock could be called a "non-gene" because it has nothing to do with genetics, but it doesn't regulate the expression of genes in a knowledge laden way.

Alan

Rami Rustom

unread,
Jan 2, 2012, 1:54:30 AM1/2/12
to Beginning of Infinity
On Jan 2, 12:35 am, Alan Forrester
Ya it was a bit confusing when I wrote it; especially since I said
*non-gene* and *genetic* material in the same sentence. But your
argument cleared it up. Maybe they should be called f-genes and m-
genes; F for factory and M for management.

--Rami

Stephen Push

unread,
Jan 2, 2012, 5:41:46 PM1/2/12
to Beginning of Infinity
On Jan 1, 3:31 pm, Rami Rustom <ramir...@gmail.com> wrote:

> 1. Frogs learn (hardwired) implicit rules.
> 2. Monkeys do that plus they learn (softwired) implicit situations and rules.
> 3. Pre-language Humans do that plus they learn (softwired) logics implicitly.
> 4. Post-language Humans do that plus they learn (softwired) logics explicitly.
>
> So consider this question: After millions of years of evolution from
> basic animals into more complex animals, and then into even more
> complex animals, why didn't the number of genes increase? The answer
> is DD's principle of universality.

I agree with your example. But the universality of DNA/RNA is even
more impressive than that. The same four-base code is used to encode
everything from bacteria to slime molds, oak trees, worms, humans,
etc.

> 3. Then the non-gene knowledge reached a level where softwired
> implicit 2nd order knowledge creation was occurring. This resulted in
> humans; and evolution stopped.

Humans are still undergoing biological evolution, but it has been
supplemented by cultural evolution. I think when DD said evolution
stopped, he was referring to the four-base nucleic acid code. Once
that code appeared, it was so successful that it gave rise to all of
the organisms on Earth. Evolution probably continued with respect to
gene regulation, DNA repair, and other refinements, but the basic code
has not changed in billions of years. In DD's terminology, the
genetic code has tremendous "reach."

Steve

Rami Rustom

unread,
Jan 2, 2012, 6:11:26 PM1/2/12
to Beginning of Infinity
On Jan 2, 4:41 pm, Stephen Push <stephen.p...@gmail.com> wrote:
> On Jan 1, 3:31 pm, Rami Rustom <ramir...@gmail.com> wrote:
>
> > 1. Frogs learn (hardwired) implicit rules.
> > 2. Monkeys do that plus they learn (softwired) implicit situations and rules.
> > 3. Pre-language Humans do that plus they learn (softwired) logics implicitly.
> > 4. Post-language Humans do that plus they learn (softwired) logics explicitly.
>
> > So consider this question: After millions of years of evolution from
> > basic animals into more complex animals, and then into even more
> > complex animals, why didn't the number of genes increase? The answer
> > is DD's principle of universality.
>
> I agree with your example.  But the universality of DNA/RNA is even
> more impressive than that.  The same four-base code is used to encode
> everything from bacteria to slime molds, oak trees, worms, humans,
> etc.

Yes. Universality happens at each stage. You've mentioned a stage
prior to the stages I mentioned.


> > 3. Then the non-gene knowledge reached a level where softwired
> > implicit 2nd order knowledge creation was occurring. This resulted in
> > humans; and evolution stopped.
>
> Humans are still undergoing biological evolution, but it has been
> supplemented by cultural evolution.

No. Biological evolution has stopped. Genes are no longer replicating
based on their fitness. Our technology is preventing it. Our
technology keeps people alive therefore genes are no longer competing
with their allele counterparts.


> I think when DD said evolution
> stopped, he was referring to the four-base nucleic acid code.  Once
> that code appeared, it was so successful that it gave rise to all of
> the organisms on Earth.

Yes. That was one level of universality.


> Evolution probably continued with respect to
> gene regulation, DNA repair, and other refinements, but the basic code
> has not changed in billions of years.

An example of universality on the gene level is the genetic code that
encodes skeletal muscle. These genes were created long after the 4-
base code and long before my stage 1 example.

I have another cool example. It has been conjectured that life at one
point in history used H2S instead of H2O. The genes to use H2O
competed with the genes to use H2S and competition revealed the
winner.

Btw, a really cool question: "Why do our bodies need oxygen?" The
answer is a physical property that oxygen has that makes it more
*useful* than any other element.


> In DD's terminology, the
> genetic code has tremendous "reach."

Yes each instance of universality has reach which is the very
definition of universality.

--Rami

stephe...@gmail.com

unread,
Jan 2, 2012, 10:36:02 PM1/2/12
to Beginning of Infinity
On Jan 2, 6:11 pm, Rami Rustom <ramir...@gmail.com> wrote:

> No. Biological evolution has stopped. Genes are no longer replicating
> based on their fitness. Our technology is preventing it. Our
> technology keeps people alive therefore genes are no longer competing
> with their allele counterparts.

The evidence doesn't support that belief. See the following two
articles:

http://www.pnas.org/content/early/2011/09/28/1104210108

http://www.livescience.com/7971-humans-evolving-brains-shrink.html

Steve

Rami Rustom

unread,
Jan 2, 2012, 10:56:59 PM1/2/12
to Beginning of Infinity
On Jan 2, 9:36 pm, "stephen.p...@gmail.com" <stephen.p...@gmail.com>
wrote:
Interesting. So there must still exist selection pressures on genes.
So it seems your explanation is correct, that biological evolution has
slowed because meme evolution has increased. This makes a lot more
sense anyway.

My idea was absolute. Absolutes are usually wrong. So in each of the
stage transitions, there was an inclining of the newer stage evolution
that caused a declining of the earlier stage evolution.

The stages (including the one you mentioned and others)...
i. Eukaryotas use 4-base code for biological knowledge for DNA
replication.
ii. (many others stages)
iii. Sentient organisms...
1. Frogs learn (hardwired) implicit rules.
2. Monkeys do that plus they learn (softwired) implicit situations and
rules.
3. Pre-language Humans do that plus they learn (softwired) logics
implicitly. (Sapience)
4. Post-language Humans do that plus they learn (softwired) logics
explicitly.
5. Will there be another level of universality?

What do you think?

--Rami

Elliot Temple

unread,
Jan 2, 2012, 11:31:34 PM1/2/12
to beginning-...@googlegroups.com

On Jan 2, 2012, at 7:36 PM, stephe...@gmail.com wrote:

> On Jan 2, 6:11 pm, Rami Rustom <ramir...@gmail.com> wrote:
>
>> No. Biological evolution has stopped. Genes are no longer replicating
>> based on their fitness. Our technology is preventing it. Our
>> technology keeps people alive therefore genes are no longer competing
>> with their allele counterparts.
>
> The evidence doesn't support that belief. See the following two
> articles:
>
> http://www.pnas.org/content/early/2011/09/28/1104210108

Says

> Here we study life-history variation in an insular preindustrial French-Canadian population


If they are intentionally studying insular pre-modern people, how is that relevant to Rami's claim about modern *technology* preventing continued biological evolution?

> http://www.livescience.com/7971-humans-evolving-brains-shrink.html

Says

> Comprehensive scans of the human genome reveal that hundreds of our genes show evidence of changes during the past 10,000 years of human evolution.


How are changes since 10,000 years ago relevant to Rami's claims about modern technology, which has only been around for, say, 100 years?

Where I'd want to start is by considering: what are the selection pressures and how are they met? We have to come up with explanations.

For example, in modern technological society you do not need to be physically strong to have children. And the number of children you have is primarily up to you, not up to the sort of fitness that would have mattered 100,000 years ago.

So if there was evidence we are stronger than we used to be -- which would not surprise me -- I would look at things like diet, leisure time for exercise, cultural preferences and improved training methods. Not genes since there isn't a compelling explanation of why there'd be a substantial selection pressure on genes for this.

Does that methodology make sense to others?


-- Elliot Temple
http://elliottemple.com/

stephe...@gmail.com

unread,
Jan 2, 2012, 11:35:47 PM1/2/12
to Beginning of Infinity
On Jan 2, 10:56 pm, Rami Rustom <ramir...@gmail.com> wrote:

> The stages (including the one you mentioned and others)...
> i.  Eukaryotas use 4-base code for biological knowledge for DNA
> replication.
> ii. (many others stages)
> iii. Sentient organisms...
> 1. Frogs learn (hardwired) implicit rules.
> 2. Monkeys do that plus they learn (softwired) implicit situations and
> rules.
> 3. Pre-language Humans do that plus they learn (softwired) logics
> implicitly.  (Sapience)
> 4. Post-language Humans do that plus they learn (softwired) logics
> explicitly.
> 5. Will there be another level of universality?

In one sense, I think you are right. Human learning is more universal
than a frog's because it can solve a wider range of problems. But I
find your progression anthropocentric and teleological. A frog's
learning is well-suited to its ecological niche. In some situations,
hard-wired responses are superior to soft-wired.

I think the universality of the DNA code is of a different sort than
human intelligence. Human intelligence is but one manifestation of
the DNA code -- and a relatively recent one at that. It remains to be
seen whether human intelligence will enable long-term survival. DNA
has been around for billions of years. The human lineage diverged
from the other apes just 8 million years ago.

Steve

stephe...@gmail.com

unread,
Jan 3, 2012, 12:22:51 AM1/3/12
to Beginning of Infinity
On Jan 2, 11:31 pm, Elliot Temple <c...@curi.us> wrote:

> If they are intentionally studying insular pre-modern people, how is that relevant to Rami's claim about modern
> *technology* preventing continued biological evolution?

Pre-industrial is not the same as pre-modern. And human technology
long predates the Industrial Revolution.

> How are changes since 10,000 years ago relevant to Rami's claims about modern technology, which has only been
> around for, say, 100 years?

Rami's claim is that human biological evolution has stopped. The
evidence available so far suggests otherwise. I don't know if we can
accurately measure evolution over the past 100 years in humans, given
our relatively long generation time.

Current technology has changed the selection pressures, but I doubt
that it has eliminated natural selection.

Steve

Rami Rustom

unread,
Jan 2, 2012, 11:47:05 PM1/2/12
to Beginning of Infinity
On Jan 2, 10:35 pm, "stephen.p...@gmail.com" <stephen.p...@gmail.com>
wrote:
> On Jan 2, 10:56 pm, Rami Rustom <ramir...@gmail.com> wrote:
>
> > The stages (including the one you mentioned and others)...
> > i.  Eukaryotas use 4-base code for biological knowledge for DNA
> > replication.
> > ii. (many others stages)
> > iii. Sentient organisms...
> > 1. Frogs learn (hardwired) implicit rules.
> > 2. Monkeys do that plus they learn (softwired) implicit situations and
> > rules.
> > 3. Pre-language Humans do that plus they learn (softwired) logics
> > implicitly.  (Sapience)
> > 4. Post-language Humans do that plus they learn (softwired) logics
> > explicitly.
> > 5. Will there be another level of universality?
>
> In one sense, I think you are right.  Human learning is more universal
> than a frog's because it can solve a wider range of problems.  But I
> find your progression anthropocentric and teleological.  A frog's
> learning is well-suited to its ecological niche.  In some situations,
> hard-wired responses are superior to soft-wired.

Yes. I didn't suggest that frogs need to become humans; which is what
I think you are saying. They are very well suited for their
environments. Thats how they evolved into frogs.


> I think the universality of the DNA code is of a different sort than
> human intelligence.  Human intelligence is but one manifestation of
> the DNA code -- and a relatively recent one at that.  It remains to be
> seen whether human intelligence will enable long-term survival. DNA
> has been around for billions of years.  The human lineage diverged
> from the other apes just 8 million years ago.

Hmm. BoI explains that knowledge will take us anywhere and everywhere.
Lack of knowledge will be the thing that kills us.

So yes its a matter of when the next big calamity will come and
whether we can create enough knowledge to stop our extinction.

But its not the human intelligence that will or will not cause long-
term survival. Its the emergent property that arises from human
intelligence that would cause it, i.e. universal explanatory
knowledge.

--Rami

stephe...@gmail.com

unread,
Jan 3, 2012, 1:14:23 AM1/3/12
to Beginning of Infinity
On Jan 2, 11:47 pm, Rami Rustom <ramir...@gmail.com> wrote:

> Hmm. BoI explains that knowledge will take us anywhere and everywhere.
> Lack of knowledge will be the thing that kills us.

Not necessarily. There is a family of fishes in Africa, mormyrids,
have use weak electric discharges to navigate and communicate. Most
members of this family have large brains (the brain-to-body size ratio
is greater than in humans), which probably evolved to process the
electric organ discharge information. But big brains are expensive:
they need lots of energy and lots of oxygen. Several mormyrid species
that live in oxygen-poor waters have evolved smaller brains.

At some time in the future, big brains could become a liability to our
descendants, who might evolve smaller brains to survive.

I'm not suggesting that will actually happen. My point is that none
of us know what the future will hold.

Steve

Rami Rustom

unread,
Jan 3, 2012, 8:11:26 AM1/3/12
to Beginning of Infinity
On Jan 3, 12:14 am, "stephen.p...@gmail.com" <stephen.p...@gmail.com>
wrote:
Three things to consider:

1* Your proposed future does not reconcile with the principle of
universality. The human mind's ability to create universal explanatory
knowledge is one of the stages of universality that biology has
reached. Universality does not go in the reverse direction. In this
specific case of stage transition, your idea suggests that we could
devolve from stage 4 back to 3. But the principle of universality
would stop it. Our technology, which is based on explicit 2nd order
knowledge, would provide the stopping power.

2* Regarding the organisms you mentioned that evolved smaller brains,
they did not reach a level of universality. There was no stage
transition. So the principle of universality does not hold for them.

3* Regarding the bigger/smaller brain idea, bigger brains doesn't mean
greater intelligence. Intelligence is about the number of possible
connections between neurons and how those connections form. The
Neanderthals had bigger brains that us. But our brains are structured
better. Therefore we are more intelligent; they died while we lived.

--Rami

stephe...@gmail.com

unread,
Jan 3, 2012, 11:55:51 AM1/3/12
to Beginning of Infinity
On Jan 3, 8:11 am, Rami Rustom <ramir...@gmail.com> wrote:

> 1* Your proposed future does not reconcile with the principle of
> universality. The human mind's ability to create universal explanatory
> knowledge is one of the stages of universality that biology has
> reached. Universality does not go in the reverse direction. In this
> specific case of stage transition, your idea suggests that we could
> devolve from stage 4 back to 3. But the principle of universality
> would stop it. Our technology, which is based on explicit 2nd order
> knowledge, would provide the stopping power.

There is no guarantee that evolution -- either biological or cultural
-- won't reverse direction in the sense of going from more complex to
less complex. The dinosaurs died out. Much of Greek learning was
lost to European culture for centuries. Even "universal" properties
could be lost. Even the four-base DNA code could die out and/or be
replaced by something more fit, in the Darwinian sense.

> 2* Regarding the organisms you mentioned that evolved smaller brains,
> they did not reach a level of universality. There was no stage
> transition. So the principle of universality does not hold for them.

We don't know that. These fishes display complex navigation and
communication behaviors. I submit that your dismissal of their
abilities is based more on anthropocentric bias than knowledge of
their biology and psychology.

> 3* Regarding the bigger/smaller brain idea, bigger brains doesn't mean
> greater intelligence. Intelligence is about the number of possible
> connections between neurons and how those connections form. The
> Neanderthals had bigger brains that us. But our brains are structured
> better. Therefore we are more intelligent; they died while we lived.

Yes, bigger isn't always better. But in evolution, "smarter" isn't
always better either.

How do you know that modern humans were smarter that Neanderthals?
Perhaps they had superior culture. Perhaps they were more ruthless.
Perhaps they were lucky.

And the Neanderthals aren't really gone. Unless you are African, you
probably carry some of their genes.

Steve

Rami Rustom

unread,
Jan 3, 2012, 6:27:33 PM1/3/12
to Beginning of Infinity
On Jan 3, 10:55 am, "stephen.p...@gmail.com" <stephen.p...@gmail.com>
wrote:
> On Jan 3, 8:11 am, Rami Rustom <ramir...@gmail.com> wrote:
>
> > 1* Your proposed future does not reconcile with the principle of
> > universality. The human mind's ability to create universal explanatory
> > knowledge is one of the stages of universality that biology has
> > reached. Universality does not go in the reverse direction. In this
> > specific case of stage transition, your idea suggests that we could
> > devolve from stage 4 back to 3. But the principle of universality
> > would stop it. Our technology, which is based on explicit 2nd order
> > knowledge, would provide the stopping power.
>
> There is no guarantee that evolution -- either biological or cultural
> -- won't reverse direction in the sense of going from more complex to
> less complex.  The dinosaurs died out.

When a species goes extinct, it is not experiencing a reverse in
evolution. It is only dying. Each species is evolving on its own.
Extinction is not devolution.


> Much of Greek learning was
> lost to European culture for centuries.

Yes happens. But I think it only happens with memes, not genes. But
I'm at a loss to explain why.

This is a very interesting problem...


> Even "universal" properties
> could be lost. Even the four-base DNA code could die out and/or be
> replaced by something more fit, in the Darwinian sense.

I disagree. Only the single-celled organisms could have the potential
to change their DNA code structure into something besides the 4-base
system. An organism composed of trillions of cells could not do this.


> > 2* Regarding the organisms you mentioned that evolved smaller brains,
> > they did not reach a level of universality. There was no stage
> > transition. So the principle of universality does not hold for them.
>
> We don't know that.  These fishes display complex navigation and
> communication behaviors.  I submit that your dismissal of their
> abilities is based more on anthropocentric bias than knowledge of
> their biology and psychology.

Ok we don't know it. But it follows, i.e. reaches, from DD's
universality principle.


> > 3* Regarding the bigger/smaller brain idea, bigger brains doesn't mean
> > greater intelligence. Intelligence is about the number of possible
> > connections between neurons and how those connections form. The
> > Neanderthals had bigger brains that us. But our brains are structured
> > better. Therefore we are more intelligent; they died while we lived.
>
> Yes, bigger isn't always better.  But in evolution, "smarter" isn't
> always better either.
>
> How do you know that modern humans were smarter that Neanderthals?
> Perhaps they had superior culture.  Perhaps they were more ruthless.
> Perhaps they were lucky.

Neanderthals lived in small tribes and their tribes didn't interact.
We lived in small tribes but our tribes interacted and traded tools.
So memes were traded too. Neanderthals weren't learning memes as
effectively because they weren't as social as we were.


> And the Neanderthals aren't really gone. Unless you are African, you probably carry some of their genes.

Yes I know of at least one instance where a child skeleton (in present
day Israel) was found that was said to have Neanderthal and human
features. So yes some of their genes are in us.

--Rami

stephe...@gmail.com

unread,
Jan 4, 2012, 7:45:29 AM1/4/12
to Beginning of Infinity
On Jan 3, 6:27 pm, Rami Rustom <ramir...@gmail.com> wrote:
> On Jan 3, 10:55 am, "stephen.p...@gmail.com" <stephen.p...@gmail.com>

> When a species goes extinct, it is not experiencing a reverse in
> evolution. It is only dying. Each species is evolving on its own.
> Extinction is not devolution.

That is correct. But there is no law that says "more complex" or
"smarter" species must prevail. In our anthropocentric view, we like
to think of ourselves as the pinnacle of evolution. Sixty-five
million years ago, an observer might have concluded that the dinosaurs
were the pinnacle of evolution. Six-five million years from now, the
Earth may be dominated by slime molds or by dumber descendants of
ourselves. Natural selection favors organisms that are adapted to
their environment, not necessarily smart or complex organisms.

The one possible exception to this scenario is that, in learning to
understand evolution and genetics, we might be able to control the
process to improve our own chances of survival. We can already
control genetic diseases to some extent. It remains to be seen
whether tinkering with the evolutionary process will be a good thing
in the long run.

> Yes happens. But I think it only happens with memes, not genes. But
> I'm at a loss to explain why.

Genetic information is lost all the time. That is one of the great
tragedies of the recent rapid decline in biodiversity caused by
habitat loss and other factors..

> I disagree. Only the single-celled organisms could have the potential
> to change their DNA code structure into something besides the 4-base
> system. An organism composed of trillions of cells could not do this.

I'm not saying that DNA-based organisms would switch to a new code.
But a different code might arise, either de novo or introduced from an
alien world.

But perhaps you are right: A universal genetic code might generally
win out in competition with a non-universal code. That might be an
interesting idea to pursue with a computer model.

> Ok we don't know it. But it follows, i.e. reaches, from DD's
> universality principle.

I don't see how application of the universality principle can reveal
the abilities of mormyrid fishes. That seems like an empirical
question to me.

> Yes I know of at least one instance where a child skeleton (in present
> day Israel) was found that was said to have Neanderthal and human
> features. So yes some of their genes are in us.

There is more evidence than that. Neanderthal DNA samples have been
obtained from bones and compared to modern human DNA. The two groups
interbred when modern humans migrated out of Africa. All human
populations outside Africa carry Neanderthal DNA.

Steve

Rami Rustom

unread,
Jan 4, 2012, 11:52:22 AM1/4/12
to Beginning of Infinity
On Jan 4, 6:45 am, "stephen.p...@gmail.com" <stephen.p...@gmail.com>
wrote:
> On Jan 3, 6:27 pm, Rami Rustom <ramir...@gmail.com> wrote:
>
> > On Jan 3, 10:55 am, "stephen.p...@gmail.com" <stephen.p...@gmail.com>
> > When a species goes extinct, it is not experiencing a reverse in
> > evolution. It is only dying. Each species is evolving on its own.
> > Extinction is not devolution.
>
> That is correct.  But there is no law that says "more complex" or
> "smarter" species must prevail.

I don't know what you mean by *must prevail*. Could you define that?

> In our anthropocentric view, we like
> to think of ourselves as the pinnacle of evolution. Sixty-five
> million years ago, an observer might have concluded that the dinosaurs
> were the pinnacle of evolution. Six-five million years from now, the
> Earth may be dominated by slime molds or by dumber descendants of
> ourselves. Natural selection favors organisms that are adapted to
> their environment, not necessarily smart or complex organisms.

The only way that universal creators like ourselves would go extinct
is if a huge calamity happened and we didn't adapt to it because we
didn't yet have the right technologies to prevent our extinction. If
this occurred, then yes slime molds or another dumber descendant of
ourselves could become the *dominant* species on Earth.

Btw, what do you mean by *dominant*? Do you mean by population? If so,
ants dominate us now. Or is it something else?

I think we're talking past each other. I didn't mention dominance in
my argument.


> The one possible exception to this scenario is that, in learning to
> understand evolution and genetics, we might be able to control the
> process to improve our own chances of survival.  We can already
> control genetic diseases to some extent.  It remains to be seen
> whether tinkering with the evolutionary process will be a good thing
> in the long run.
>
> > Yes happens. But I think it only happens with memes, not genes. But
> > I'm at a loss to explain why.
>
> Genetic information is lost all the time.  That is one of the great
> tragedies of the recent rapid decline in biodiversity caused by
> habitat loss and other factors..

Again I think we're talking past each other. You are talking about
gene loss due to extinction. I'm referring to gene loss in a specific
species; a species that doesn't go extinct.


> > I disagree. Only the single-celled organisms could have the potential
> > to change their DNA code structure into something besides the 4-base
> > system. An organism composed of trillions of cells could not do this.
>
> I'm not saying that DNA-based organisms would switch to a new code.
> But a different code might arise, either de novo or introduced from an
> alien world.

Sure but that would be a completely different species; which is
irrelevant to my argument (another case of talking past each other). I
thought you were saying that an existing 4-base code multi-cell
organism could experience a selective pressure that would change its 4-
base code system to something else. This I disagree with because it
doesn't reconcile with Deutsch's universality principle.


> But perhaps you are right:  A universal genetic code might generally
> win out in competition with a non-universal code.  That might be an
> interesting idea to pursue with a computer model.

I don't' know what you mean. If a single-cell organism experienced a
selective pressure that caused its 4-base system to change, this would
also be a code that would be dictated by Deutsch's universality
principle. A non-universal code is not possible.


> > Ok we don't know it. But it follows, i.e. reaches, from DD's
> > universality principle.
>
> I don't see how application of the universality principle can reveal
> the abilities of mormyrid fishes.  That seems like an empirical
> question to me.

I don't know what you mean by *that seems like an empirical question*.
Could you rephrase?


> > Yes I know of at least one instance where a child skeleton (in present
> > day Israel) was found that was said to have Neanderthal and human
> > features. So yes some of their genes are in us.
>
> There is more evidence than that.  Neanderthal DNA samples have been
> obtained from bones and compared to modern human DNA. The two groups
> interbred when modern humans migrated out of Africa.  All human
> populations outside Africa carry Neanderthal DNA.

So cool!

So is there evidence that those gene pools didn't mix with African
gene pools?

--Rami

stephe...@gmail.com

unread,
Jan 4, 2012, 3:52:38 PM1/4/12
to Beginning of Infinity
On Jan 4, 11:52 am, Rami Rustom <ramir...@gmail.com> wrote:

> I don't know what you mean by *must prevail*. Could you define that?

By prevail I mean pass genes into future generations. Although the
overall trend in evolution has been more complex organisms evolving
from less complex, the opposite has happened from time to time. But I
won't belabor this point unless you think it is still on topic.

> Again I think we're talking past each other. You are talking about
> gene loss due to extinction. I'm referring to gene loss in a specific
> species; a species that doesn't go extinct.

That's a tricky question because of the ambiguity in trying to
identify different species over time. Clearly some genes (alleles to
be more precise) are extinguished from populations. But is the former
population the same species as the current one? That's a somewhat
arbitrary determination, because the most common species concept, the
biological species concept, depends on whether two species
interbreed. Again, I'll drop this line of discussion if you think
it's getting off topic.

> Sure but that would be a completely different species; which is
> irrelevant to my argument (another case of talking past each other). I
> thought you were saying that an existing 4-base code multi-cell
> organism could experience a selective pressure that would change its 4-
> base code system to something else. This I disagree with because it
> doesn't reconcile with Deutsch's universality principle.

I don't know if it violates the the universality principle, but it
seems like a physical impossibility. Whether it's the best possible
code or not, all current life on Earth is irrevocably tied to this
code.

> I don't' know what you mean. If a single-cell organism experienced a
> selective pressure that caused its 4-base system to change, this would
> also be a code that would be dictated by Deutsch's universality
> principle. A non-universal code is not possible.

Not according to Deutsch. He postulates non-universal codes that
arose early in the origin of life but didn't last. Eventually,
according to Deutsch, at least one of these codes developed
"universality" and became the DNA/RNA code we have today. Perhaps the
DNA/RNA code out-competed the other codes for limited resources.
That's an idea that could be tested in a laboratory or a computer
model. (The experiment wouldn't indicate whether it actually happened
-- just whether it is possible.)

> I don't know what you mean by *that seems like an empirical question*.
> Could you rephrase?

The best way to determine the properties of the navigation and
communication systems of mormyrid fishes is through observation and
experimentation. Applying a principle like universality will not
provide any reliable information -- although perhaps it could help
generate some interesting hypotheses to test.

> So is there evidence that those gene pools didn't mix with African
> gene pools?

Pure sub-Saharan African populations don't carry the Neanderthal genes
because the prehistoric migration was away from sub-Saharan Africa.
But that is changing now. In the last few centuries, Europeans,
Asians, and other non-Africans have migrated to Africa, while people
of African descent can be found throughout the world today. Mixed
race children will inherit some Neanderthal genes. If the Neanderthal
genes persist, I would except them to be worldwide someday.

Steve

Rami Rustom

unread,
Jan 4, 2012, 5:26:46 PM1/4/12
to Beginning of Infinity
On Jan 4, 2:52 pm, "stephen.p...@gmail.com" <stephen.p...@gmail.com>
wrote:
> On Jan 4, 11:52 am, Rami Rustom <ramir...@gmail.com> wrote:
>
> > I don't know what you mean by *must prevail*. Could you define that?
>
> By prevail I mean pass genes into future generations.  Although the
> overall trend in evolution has been more complex organisms evolving
> from less complex, the opposite has happened from time to time.  But I
> won't belabor this point unless you think it is still on topic.

Hi Steve, I'm sorry if I've given the impression that your point was
not on topic. All I meant was that we were talking about 2 different
points. I think that both points are on topic and that we should
continue discussing them until we reach agreement. :)

Btw, I think all points are always on topic. All knowledge is
connected, either directly or indirectly. I don't like the idea of
separating fields of study, just like David and Elliot. Any tangent
could be interesting and for me, if we have disagreement, then its
interesting.

On another note, I think that its possible that all of the
disagreements that still exist in this thread are actually just one
disagreement. But since we can't know this yet, I'll continue to treat
each one separately.

So back to the subject, could you provide a criticism in the form of
an example of a case of devolution?

If you are talking about the fish example, can you tell me whether
that example involves going in the reverse direction from one emergent
stage to a previous emergent stage?
-- If not, then we're not talking about the same thing. So in this
case, maybe you are saying that devolution can occur within a stage?
If so, I agree on this because universality doesn't hold between stage
transitions.


> > Again I think we're talking past each other. You are talking about
> > gene loss due to extinction. I'm referring to gene loss in a specific
> > species; a species that doesn't go extinct.
>
> That's a tricky question because of the ambiguity in trying to
> identify different species over time.  Clearly some genes (alleles to
> be more precise) are extinguished from populations.  But is the former
> population the same species as the current one?  That's a somewhat
> arbitrary determination, because the most common species concept, the
> biological species concept, depends on whether two species
> interbreed.  Again, I'll drop this line of discussion if you think
> it's getting off topic.

Think of a line of species instead of just one species. A line of
species, which has reached a stage of universality, will not devolve
to the previous stage. This follows from the universality principle.



> > Sure but that would be a completely different species; which is
> > irrelevant to my argument (another case of talking past each other). I
> > thought you were saying that an existing 4-base code multi-cell
> > organism could experience a selective pressure that would change its 4-
> > base code system to something else. This I disagree with because it
> > doesn't reconcile with Deutsch's universality principle.
>
> I don't know if it violates the the universality principle, but it
> seems like a physical impossibility.  Whether it's the best possible
> code or not, all current life on Earth is irrevocably tied to this
> code.

On this point, it seems that the only possible way to move forward to
reach agreement is for you to reread the universality chapter in BoI
in order to form a criticism (since I created the conjecture).

What chapter was that?



> > I don't' know what you mean. If a single-cell organism experienced a
> > selective pressure that caused its 4-base system to change, this would
> > also be a code that would be dictated by Deutsch's universality
> > principle. A non-universal code is not possible.
>
> Not according to Deutsch.  He postulates non-universal codes that
> arose early in the origin of life but didn't last.  Eventually,
> according to Deutsch, at least one of these codes developed
> "universality" and became the DNA/RNA code we have today.  Perhaps the
> DNA/RNA code out-competed the other codes for limited resources.
> That's an idea that could be tested in a laboratory or a computer
> model.  (The experiment wouldn't indicate whether it actually happened
> -- just whether it is possible.)

Ah let me clarify. In my previous argument, I was assuming that
another stage of universality had been reached.

In your example, yes a non-universal code could exist while the
organism was sort of trying out new codes (very bad description I
know).



> > I don't know what you mean by *that seems like an empirical question*.
> > Could you rephrase?
>
> The best way to determine the properties of the navigation and
> communication systems of mormyrid fishes is through observation and
> experimentation.  Applying a principle like universality will not
> provide any reliable information -- although perhaps it could help
> generate some interesting hypotheses to test.

Oh I see what you mean. But we don't need to do this just yet. The
above argument should suffice.

stephe...@gmail.com

unread,
Jan 5, 2012, 1:06:00 AM1/5/12
to Beginning of Infinity
On Jan 4, 5:26 pm, Rami Rustom <ramir...@gmail.com> wrote:

> So back to the subject, could you provide a criticism in the form of
> an example of a case of devolution?

I don't like the term "devolution" because it implies that organisms
that lose some structures or functions are degenerate forms. This
term arises from the teleological misconception that evolution always
progresses toward greater complexity or perfection. In fact, all
species -- simple and complex -- are adapted to their respective
environments.

Some additional examples of evolution from complex to simple:

1) Sexual reproduction was one of the major transitions in
evolution. But some sexually reproducing species have reverted to
asexual reproduction.

2) Various cave-dwelling species have lost their eyes.

3) Dogs have lost many of the social and pup-rearing behaviors
displayed by their wolf ancestors. This example fulfills your request
for a complex-to-simple change within the same species. Since wolves
and dogs interbreed, they are the same species, according to the
biological species concept used by zoologists.

4) Several species of frogs have lost the tadpole stage of their life
cycles.

5) Some scientists believe that prokaryotes (organisms such as
bacteria with simple cells) may have evolved from eukaryotes
(organisms with complex cells like ours). See
http://www.msnbc.msn.com/id/12853798/ns/technology_and_science-science/t/can-evolution-make-things-less-complicated/#.TwU4LFYiFXt
(This article also mentions two other examples of complex-to-simple
evolution: the malaria parasite and baker's yeast.)

> If you are talking about the fish example, can you tell me whether
> that example involves going in the reverse direction from one emergent
> stage to a previous emergent stage?

I'm not sure what you mean by an emergent stage. But if you are
looking for major transitions in evolution, I would point you to the
sexual/asexual reproduction example (#1 above). If the hypothesis in
number 5 above is correct, that would be another example, since the
eukaryote/prokaryote transition is also a major one.

Other than the DNA code, can you give me some examples of jumps to
universality (or emergent stages) in biological evolution?

> On this point, it seems that the only possible way to move forward to
> reach agreement is for you to reread the universality chapter in BoI
> in order to form a criticism (since I created the conjecture).
>
> What chapter was that?

Chapter 6.

Steve

David Deutsch

unread,
Jan 5, 2012, 6:27:07 AM1/5/12
to beginning-...@googlegroups.com
On 5 Jan 2012, at 6:06am, stephe...@gmail.com wrote:

> I don't like the term "devolution" because it implies that organisms
> that lose some structures or functions are degenerate forms. This
> term arises from the teleological misconception that evolution always
> progresses toward greater complexity or perfection.

The term 'devolution' implies only that evolution *sometimes* produces greater 'complexity or perfection' (or, more precisely, knowledge, or progress), and sometimes goes the other way.

> In fact, all
> species -- simple and complex -- are adapted to their respective
> environments.

All? Is a species that is about to go extinct adapted to its environment? Is a species that is rapidly evolving new adaptations, adapted to its environment? Is our species adapted to life in the Great Rift Valley? Evidently those species all have *some adaptations* to their environments, but to make your case that 'devolution' is a misleading term, you would need to claim that all species are about *equally well* adapted to their environments.

-- David Deutsch

Rami Rustom

unread,
Jan 5, 2012, 8:16:24 AM1/5/12
to Beginning of Infinity
On Jan 5, 12:06 am, "stephen.p...@gmail.com" <stephen.p...@gmail.com>
wrote:
> On Jan 4, 5:26 pm, Rami Rustom <ramir...@gmail.com> wrote:
>
> > So back to the subject, could you provide a criticism in the form of
> > an example of a case of devolution?
>
> I don't like the term "devolution" because it implies that organisms
> that lose some structures or functions are degenerate forms.  This
> term arises from the teleological misconception that evolution always
> progresses toward greater complexity or perfection.  In fact, all
> species -- simple and complex -- are adapted to their respective
> environments.
>
> Some additional examples of evolution from complex to simple:
>
> 1)  Sexual reproduction was one of the major transitions in
> evolution.  But some sexually reproducing species have reverted to
> asexual reproduction.
>
> 2)  Various cave-dwelling species have lost their eyes.
>
> 3)  Dogs have lost many of the social and pup-rearing behaviors
> displayed by their wolf ancestors.  This example fulfills your request
> for a complex-to-simple change within the same species.  Since wolves
> and dogs interbreed, they are the same species, according to the
> biological species concept used by zoologists.
>
> 4)  Several species of frogs have lost the tadpole stage of their life
> cycles.
>
> 5)  Some scientists believe that prokaryotes (organisms such as
> bacteria with simple cells) may have evolved from eukaryotes
> (organisms with complex cells like ours).  Seehttp://www.msnbc.msn.com/id/12853798/ns/technology_and_science-scienc...
> (This article also mentions two other examples of complex-to-simple
> evolution: the malaria parasite and baker's yeast.)
>
> > If you are talking about the fish example, can you tell me whether
> > that example involves going in the reverse direction from one emergent
> > stage to a previous emergent stage?
>
> I'm not sure what you mean by an emergent stage.  But if you are
> looking for major transitions in evolution, I would point you to the
> sexual/asexual reproduction example (#1 above).  If the hypothesis in
> number 5 above is correct, that would be another example, since the
> eukaryote/prokaryote transition is also a major one.

An *emergent stage* is one that has *emergent properties* that the
previous did not have. Deutsch defines *emergent properties* in BoI.

Your *major transition* is not equivalent to my *stage transition*.
Stage transitions only occur when a level of universality is reached.
And once that level is reached, the line of species can not devolve
back to a previous stage.


> Other than the DNA code, can you give me some examples of jumps to
> universality (or emergent stages) in biological evolution?

All of the stages I mentioned are examples. I'll include them:
i. Single-celled organisms use 4-base code for biological knowledge
for DNA
replication.
ii. (many others stages)
iii. Sentient organisms...
1. Frogs learn (hardwired) implicit rules.
2. Monkeys do that plus they learn (softwired) implicit situations and
rules.
3. Pre-language Humans do that plus they learn (softwired) logics
implicitly. (Sapience)
4. Post-language Humans do that plus they learn (softwired) logics
explicitly.


> > On this point, it seems that the only possible way to move forward to
> > reach agreement is for you to reread the universality chapter in BoI
> > in order to form a criticism (since I created the conjecture).
>
> > What chapter was that?
>
> Chapter 6.

Ok so I did read that chapter before starting this thread. I wasn't
sure.

--Rami

Rami Rustom

unread,
Jan 5, 2012, 8:19:52 AM1/5/12
to Beginning of Infinity
On Jan 5, 5:27 am, David Deutsch <david.deut...@qubit.org> wrote:
One more thing to note is the definition of *complexity* provided in
the original post.

First I'll define *complex*.
*Frogs minds have *hardwired rules*; these are implicitly known.
*Monkey minds have that plus *softwired situations* and *softwired
rules*
*Human minds have that plus *softwired logics*; which can be
implicitly or explicitly learned.

So I was only speaking of the mind. All of your complexity examples
are not about the mind.

--Rami

//////////////

Hi David :)

I'm glad you chimed in because I was going to drop the devolution
term. It seems I give up to easily.

And the argument you provided clearly showed me that I didn't really
understand Steve's last post.

Ambiguity is so hard to notice, especially for a novice reader/writer.
And so many ideas are lost because of it.

Thanks :)

--Rami

stephe...@gmail.com

unread,
Jan 5, 2012, 9:36:35 PM1/5/12
to Beginning of Infinity
On Jan 5, 6:27 am, David Deutsch <david.deut...@qubit.org> wrote:

> The term 'devolution' implies only that evolution *sometimes* produces greater 'complexity or perfection' (or, more precisely, knowledge, or progress), and sometimes goes the other way.

That’s what I don’t like about the term. “Perfection” and “progress”
are teleological.

> All? Is a species that is about to go extinct adapted to its environment? Is a species that is rapidly evolving new adaptations, adapted to its environment? Is our species adapted to life in the Great Rift Valley? Evidently those species all have *some adaptations* to their environments, but to make your case that 'devolution' is a misleading term, you would need to claim that all species are about *equally well* adapted to their environments.

Actually, all species are about equally well-adapted to previous
generations’ environments. When environments change, species are less
well-adapted and their less-fit genotypes decline while their more-fit
genotypes increase in frequency. The species that is about to go
extinct was well-adapted to previous generations’ environments, but
it’s not devolving, it’s failing to evolve quickly enough to keep pace
with a rapidly changing environment. The species that is changing
rapidly may become more complex, less complex, or maintain the same
complexity – whatever it takes to adapt to the new environment.

Steve

Rami Rustom

unread,
Jan 6, 2012, 5:15:39 AM1/6/12
to Beginning of Infinity
On Jan 5, 8:36 pm, "stephen.p...@gmail.com" <stephen.p...@gmail.com>
wrote:
> On Jan 5, 6:27 am, David Deutsch <david.deut...@qubit.org> wrote:
>
> > The term 'devolution' implies only that evolution *sometimes* produces greater 'complexity or perfection' (or, more precisely, knowledge, or progress), and sometimes goes the other way.
>
> That’s what I don’t like about the term.  “Perfection” and “progress”
> are teleological.

I don't think that my argument suggested "perfection".


> > All? Is a species that is about to go extinct adapted to its environment? Is a species that is rapidly evolving new adaptations, adapted to its environment? Is our species adapted to life in the Great Rift Valley? Evidently those species all have *some adaptations* to their environments, but to make your case that 'devolution' is a misleading term, you would need to claim that all species are about *equally well* adapted to their environments.
>
> Actually, all species are about equally well-adapted to previous
> generations’ environments.  When environments change, species are less
> well-adapted and their less-fit genotypes decline while their more-fit
> genotypes increase in frequency.  The species that is about to go
> extinct was well-adapted to previous generations’ environments, but
> it’s not devolving, it’s failing to evolve quickly enough to keep pace
> with a rapidly changing environment. The species that is changing
> rapidly may become more complex, less complex, or maintain the same
> complexity – whatever it takes to adapt to the new environment.

I agree.

But I think your conception of *complex* is different than mine, as
stated in my original post. I was referring to the mind (brain) not
the rest of biology.

So this begs the question: My explanation says that complexity in the
mind goes forward only. And your explanation says that complexity in
biology goes forward mostly but sometimes backwards too. So how do
these reconcile?

The answer is in the genes. Have you heard of some lines of species
that lost complexity, lets say eyes, and then gained them back later?
What could explain how complexity can be lost and then recovered later
in the same line of species?

Its because of the f-genes and m-genes. In the example above, f-genes
made the eyes. And then the m-genes were turned off in that line of
species. And then there were turned on again. And it was selective
pressures that caused both the off and then back on again state
transitions.

So we should define complexity more generally. Complexity is in the
genes, not the outward appearance, because some f-genes are hidden
when the m-genes are indefinitely turned off in a species.

Biology is so cool!

--Rami

stephe...@gmail.com

unread,
Jan 6, 2012, 9:04:50 PM1/6/12
to Beginning of Infinity
On Jan 6, 5:15 am, Rami Rustom <ramir...@gmail.com> wrote:

> I don't think that my argument suggested "perfection".

I don't think you did either. I was responding to David Deutsch's use
of the term "perfection."

> But I think your conception of *complex* is different than mine, as
> stated in my original post. I was referring to the mind (brain) not
> the rest of biology.

I don't think the brain and mind are different in principle from other
organs and functions. I'll do some research and see if I can find
additional examples of loss of complexity in the brain.

> So this begs the question: My explanation says that complexity in the
> mind goes forward only. And your explanation says that complexity in
> biology goes forward mostly but sometimes backwards too. So how do
> these reconcile?
>
> The answer is in the genes. Have you heard of some lines of species
> that lost complexity, lets say eyes, and then gained them back later?

Yes. Wings, eyes, digits, larvae, and tadpoles have been lost and
later re-evolved.

> What could explain how complexity can be lost and then recovered later
> in the same line of species?
>
> Its because of the f-genes and m-genes. In the example above, f-genes
> made the eyes. And then the m-genes were turned off in that line of
> species. And then there were turned on again. And it was selective
> pressures that caused both the off and then back on again state
> transitions.

What do you mean by f-genes and m-genes? I'm not familiar with those
terms.

But yes, genes can be turned on and off. It seems to me possible that
some traits could be lost by turning off certain genes and reacquired
by turning the same genes on again.

> So we should define complexity more generally. Complexity is in the
> genes, not the outward appearance, because some f-genes are hidden
> when the m-genes are indefinitely turned off in a species.

Yes, it seems possible that complexity could be lost in the phenotype
but remain dormant in the genotype. Is this just a hypothesis, or
have some studies provided evidence that this has happened?

> Biology is so cool!

It sure is!

Steve

Rami Rustom

unread,
Jan 6, 2012, 9:19:50 PM1/6/12
to Beginning of Infinity
On Jan 6, 8:04 pm, "stephen.p...@gmail.com" <stephen.p...@gmail.com>
wrote:
> On Jan 6, 5:15 am, Rami Rustom <ramir...@gmail.com> wrote:
>
> > I don't think that my argument suggested "perfection".
>
> I don't think you did either.  I was responding to David Deutsch's use
> of the term "perfection."
>
> > But I think your conception of *complex* is different than mine, as
> > stated in my original post. I was referring to the mind (brain) not
> > the rest of biology.
>
> I don't think the brain and mind are different in principle from other
> organs and functions.  I'll do some research and see if I can find
> additional examples of loss of complexity in the brain.

The brain is an organ. But the mind is a different beast. It is a
different level of emergence. The brain is biology. The mind is
consciousness. The emergent properties that exist in the mind do not
exist in the brain. And no other organ besides the brain has a higher
level of emergence resulting from it. So even in principle, the brain
is very different than the other organs.


> > So this begs the question: My explanation says that complexity in the
> > mind goes forward only. And your explanation says that complexity in
> > biology goes forward mostly but sometimes backwards too. So how do
> > these reconcile?
>
> > The answer is in thegenes. Have you heard of some lines of species
> > that lost complexity, lets say eyes, and then gained them back later?
>
> Yes.  Wings, eyes, digits, larvae, and tadpoles have been lost and
> later re-evolved.
>
> > What could explain how complexity can be lost and then recovered later
> > in the same line of species?
>
> > Its because of the f-genesand m-genes. In the example above, f-genes
> > made the eyes. And then the m-geneswere turned off in that line of
> > species. And then there were turned on again. And it was selective
> > pressures that caused both the off and then back on again state
> > transitions.
>
> What do you mean by f-genesand m-genes?  I'm not familiar with those
> terms.

I defined them in the beginning of this thread. But in the beginning
of the thread I called them genes and non-genes. Then we changed the
names to factory-genes and management-genes. The m-genes only turn on/
off the f-genes. The f-genes do the work of actually making things.


> But yes,genescan be turned on and off.  It seems to me possible that
> some traits could be lost by turning off certaingenesand reacquired
> by turning the samegeneson again.
>
> > So we should define complexity more generally. Complexity is in the
> >genes, not the outward appearance, because some f-genesare hidden
> > when the m-genesare indefinitely turned off in a species.
>
> Yes, it seems possible that complexity could be lost in the phenotype
> but remain dormant in the genotype.  Is this just a hypothesis, or
> have some studies provided evidence that this has happened?

Well it was conjectured in a documentary I saw. Don't remember the
name. Don't know any studies. If you find any though please post. :)

--Rami

stephe...@gmail.com

unread,
Jan 7, 2012, 9:00:24 AM1/7/12
to Beginning of Infinity
On Jan 6, 9:19 pm, Rami Rustom <ramir...@gmail.com> wrote:

> Well it was conjectured in a documentary I saw. Don't remember the
> name. Don't know any studies. If you find any though please post. :)

I found this review article about reverse evolution:

http://crandalllab.byu.edu/Portals/20/docs/publications/PorterTREE03.pdf

Let's discuss it if you think it's relevant.

Steve

Rami Rustom

unread,
Jan 7, 2012, 9:03:11 AM1/7/12
to Beginning of Infinity
On Jan 7, 8:00 am, "stephen.p...@gmail.com" <stephen.p...@gmail.com>
wrote:
Sure but can you quote something from the article. Looks long. I hate
to say this but I don't like reading. :)

--Rami

stephe...@gmail.com

unread,
Jan 8, 2012, 12:55:53 AM1/8/12
to Beginning of Infinity
At the risk of oversimplifying the article about reverse evolution,
I'll touch on a few of the themes:

In experiments, fruit flies that have evolved for 100 generations can
return to the original phenotype in 20 generations.

There are many cases of organisms losing traits, such as snakes losing
their limbs.

In some cases, lost traits can be regained, e.g., stick insects losing
and regaining wings. If the regaining takes place after a short
period of time, it's possible the organism could regain its original
phenotype. But it is also possible, especially if a lot time has
elapsed, that the regained trait only appears to be the same but it
based on a different genotype (i,e., convergent evolution).

There are also examples of organisms streamlining their genomes --
that is, eliminating or inactivation redundant genetic information.

Steve

Rami Rustom

unread,
Jan 8, 2012, 8:36:21 AM1/8/12
to beginning-...@googlegroups.com
On Sat, Jan 7, 2012 at 11:55 PM, stephe...@gmail.com
<stephe...@gmail.com> wrote:
> On Jan 7, 9:03 am, Rami Rustom <ramir...@gmail.com> wrote:
>
>> On Jan 7, 8:00 am, "stephen.p...@gmail.com" <stephen.p...@gmail.com>
>> wrote:
>>
>> > On Jan 6, 9:19 pm, Rami Rustom <ramir...@gmail.com> wrote:
>>
>> > > Well it was conjectured in a documentary I saw. Don't remember the
>> > > name. Don't know any studies. If you find any though please post. :)
>>
>> > I found this review article about reverse evolution:
>>
>> >http://crandalllab.byu.edu/Portals/20/docs/publications/PorterTREE03.pdf
>>
>> > Let's discuss it if you think it's relevant.
>>
>> Sure but can you quote something from the article. Looks long. I hate
>> to say this but I don't like reading. :)
>
> At the risk of oversimplifying the article about reverse evolution,
> I'll touch on a few of the themes:

So my devolution term is equivalent to *reverse evolution*. Cool.


> In experiments, fruit flies that have evolved for 100 generations

Do you mean that the researchers presented the flies with selective
pressures and that the genotype changed causing a change in phenotype,
and that it took 100 generations to cause this change?

> can return to the original phenotype in 20 generations.

If you answered yes to my above question, then yes this statement
makes sense. F-genes evolve slower. M-genes much faster.


> There are many cases of organisms losing traits, such as snakes losing
> their limbs.

Yes phenotype changes, i.e. erased, but the genotype still exists,
i.e. the f-genes are still there. And that snake line of species could
grow legs again with the proper selective pressures and some luck.

But this begs the question, do the f-genes last forever? No. And your
answer below explains why.


> In some cases, lost traits can be regained, e.g., stick insects losing
> and regaining wings.  If the regaining takes place after a short
> period of time, it's possible the organism could regain its original
> phenotype.  But it is also possible, especially if a lot time has
> elapsed, that the regained trait only appears to be the same but it
> based on a different genotype (i,e., convergent evolution).

Ah yes convergent evolution happens too. Fascinating!!!


> There are also examples of organisms streamlining their genomes --
> that is, eliminating or inactivation redundant genetic information.

Hmm. How does that work? Do they have an explanation? Or you maybe?

--Rami

Steve Push

unread,
Jan 14, 2012, 12:23:45 PM1/14/12
to Beginning of Infinity


On Jan 8, 8:36 am, Rami Rustom <ramir...@gmail.com> wrote:
> On Sat, Jan 7, 2012 at 11:55 PM, stephen.p...@gmail.com
>
>
>
>
>
>
>
>
>
> <stephen.p...@gmail.com> wrote:
> > On Jan 7, 9:03 am, Rami Rustom <ramir...@gmail.com> wrote:
>
> >> On Jan 7, 8:00 am, "stephen.p...@gmail.com" <stephen.p...@gmail.com>
> >> wrote:
>
> >> > On Jan 6, 9:19 pm, Rami Rustom <ramir...@gmail.com> wrote:
>
> >> > > Well it was conjectured in a documentary I saw. Don't remember the
> >> > > name. Don't know any studies. If you find any though please post. :)
>
> >> > I found this review article about reverse evolution:
>
> >> >http://crandalllab.byu.edu/Portals/20/docs/publications/PorterTREE03.pdf
>
> >> > Let's discuss it if you think it's relevant.
>
> >> Sure but can you quote something from the article. Looks long. I hate
> >> to say this but I don't like reading. :)
>
> > At the risk of oversimplifying the article about reverse evolution,
> > I'll touch on a few of the themes:
>
> So my devolution term is equivalent to *reverse evolution*. Cool.

I thought of you when I saw this article in Science magazine: Certain
ants carry a genotype for "supersoldiers." The supersoldier phenotype
often has not been expressed but has reappeared repeatedly. The
supersoldier genotype has been preserved for 30 million years. As you
said, even when the phenotype reverses the genetic information may
continue indefinitely.

Steve

Rami Rustom

unread,
Jan 14, 2012, 1:19:21 PM1/14/12
to beginning-...@googlegroups.com

WOW 30 million years!?!?! I guess for the ant line of species thats
not a long time. 1,100 sub species that all of that same genotype. And
currently in nature, only 6 of those 1,100 show the phenotype. I
wonder what the natural selective pressures are that turn on and off
the m-genes that cause the phenotype changes.

So the researchers used the hormone from the subspecies that already
show the phenotype to induce the m-genes to turn on in the other
subspecies.

And the researchers provided a conjecture for why the genotype wasn't lost:
> The process that produces the supersoldiers is an exaggeration of the one that produces the normal soldiers. It involves the same hormone, so it may have been impossible for the ants to get rid of the supersoldier programme entirely, without losing soldiers completely.

And the researchers also said that the induced phenotypes aren't
exactly the same as the natural ones:
> The ancestral potential has been locked in place for 35-60 million years, but when it comes back, it comes back really raw. The supersoldiers in P.obtusospinosa aren’t like the ones in P.rhea. There are differences in their behaviour and the way they’re used. It’s an interesting mix of retention and novelty. In the paper, we emphasise the sameness, but it’s actually a creative force too.

And the researchers provided a conjecture as to what selective
pressures turn on the m-genes:
> It also seems the dormant supersoldier programme spontaneously reactivates in Pheidole species from time to time. Indeed, that’s how this study started – Abouheif was studying a familiar species of ants (P.morrisi) when he suddenly found these monstrous soldiers that he had never seen before. From past experiments, we know that Pheidole sometimes produce supersoldiers if they are particularly well-nourished. “When these ancestral potentials are locked in place, there are mutations that release them at low frequencies. When natural selection needs them, it takes hold of it.

Thanks for the info. Please if you find more things like this, keep it
coming. :)

-- Rami

Reply all
Reply to author
Forward
0 new messages