The Anthropic Principle
It has recently been realised that if some of the fundamental physical constants of the universe were only slightly different then the existence of life in this universe would have been impossible (see Martin Rees's Just Six Numbers). There are many of these fortuitous coincidences which have led some to believe that the universe has been fine-tuned. Many explanations for this have been proposed: Is there some necessity for life to exist - sentient intelligent life which could observe and ponder the universe - or else the universe could not come into being? Can the conditions for life be set less rigidly? Is there a multiplicity of universes with the constants set differently in each universe? But perhaps the most famous (infamous?) interpretation of the Anthropic Principle is that we are we living in a "designer universe" (considered by Steven Weinberg here).
Critics of the Anthropic Principle dismiss such thinking by saying that human life exists in its current form purely as a result of the nature of this particular universe. If the constants were set differently then life might well not exist and we would not be around to ponder these questions. As Steven Weinberg says: "To conclude that the constants of nature have been fine-tuned by a benevolent designer would be like saying, 'Isn't it wonderful that God put us here on earth, where there's water and air and the surface gravity and temperature are so comfortable, rather than some horrid place, like Mercury or Pluto?' Where else in the solar system other than on earth could we have evolved?"
Other critics would also say that carbon-based life is not the only possible type of life: maybe an entirely different form of silicon-based life would exist, for example, if the physical constants were set differently. However, from the list of coincidences (and the particular example of the precise setting of the cosmological constant, considered here), it would appear that the majority of values which the constant could take would result in no life ever being possible, the universe either spreading too far too quickly, or else collapsing back on itself. According to the proposer of string theory Leonard Susskind: "The notorious cosmological constant is not quite zero, as it was thought to be. This is a cataclysm and the only way that we know how to make any sense of it is through the reviled and despised Anthropic Principle." (see here).
Recent discussion of the Anthropic Principle has moved away from the "designer universe" type of interpretation towards a so-called "multiverse" interpretation. This rehabilitation has seen the Anthropic Principle come in from the cold from being perceived as a slightly cranky theory on the periphery of science towards playing a central role in the latest interpretations of string theory and cosmology.
The multiverse interpretation suggests that there are a vast number of different universes (infinitely many?), the collection of universes being commonly referred-to as the multiverse. The physical constants are set to different random values in each universe. This arrangement would require no fine-tuning: we just happen to be inhabitants of a universe in which the physical constants are suitable for life.
(Max Tegmark has published a comprehensive survey of all the varied multiverse theories - see here).
The Landscape
According to string theory, elementary particles are actually one-dimensional objects: small vibrating strings. For the equations of string theory to be mathematically consistent, a string has to vibrate in 10 dimensions, which implies that six extra dimensions exist but are curled-up too small to be detected. The laws of physics in this universe would depend on the geometry of those hidden dimensions.
But the solution to the equations is not unique as so many different geometries are possible in these extra dimensions. The bundle of curled-up extra dimensions can have many different shapes (topologies), such as a sphere, a doughnut, two doughnuts joined together, and so on. The vast collection of solutions differ in that each configuration has a potential energy associated with it called the vacuum energy (basically, Einstein's cosmological constant), the energy of the spacetime when the four large dimensions are completely devoid of matter and fields.
Each solution to the equations could be taken as representing a universe with different physical constants and laws of physics. We could represent each of the 10500 possible solutions as an abstract mathematical graph, plotting the vacuum energy against the geometrical parameters. The geometry of the small dimensions would try to adjust to minimise the vacuum energy, just as a ball placed on a slope will roll downhill to a lower position. As the physical constants and laws of physics in our universe do not appear to be changing with time, we must be sitting at a minimum in the bottom of a valley. In particular, we seem to sitting at a minimum with a slightly positive vacuum energy.
Leonard Susskind has called this mathematical space The Landscape (see his Edge article or arXiv paper hep-th/0302219).
In a fascinating discussion between Leonard Susskind and Lee Smolin (see this Edge article), Smolin has no time for this "giving up on the maths" Anthropic Principle of the Landscape and suggests a Cosmological Natural Selection (CNS) process instead. According to CNS, baby universes are produced on the other side of black holes. The attraction of CNS is that the values of the physical constants might appear out of the maths as the optimal value of some kind of "fitness" measurement. The most successful universes will have most stars and black holes, and will therefore have conditions amenable to life (stars producing the heavier elements necessary for life such as carbon and oxygen).
Eternal Inflation
Leonard Susskind suggests the Landscape structure could arise within an eternally inflating universe, the structure of which was proposed by Andrei Linde (arXiv paper hep-th/0211048).
It is now widely believed that the expansion of the universe in the first moments of its existence was driven by a scalar field (also known as the inflaton field), a mysterious energy field powering the expansion of the universe. If the behaviour of the potential energy density of the scalar field is modelled then it is found that the field was slow to react to the universe's expansion ("slow-rolling down a potential energy hill" - see here), resulting in an energy density exponentially higher than would be expected. As a result, the universe experienced a period of extraordinarily rapid expansion called inflation.
Random quantum fluctuations in the scalar field,
, continue to occur due to the Heisenberg Uncertainty Principle:
If the quantum fluctuations are large enough then this can result in areas of secondary inflation (a process known as nucleation). This process can continue indefinitely. Hence, in an eternally inflating universe, "bubble" universes can bubble-up out of an eternally inflating space like bubbles in a bottle of champagne.
Due to random quantum fluctuations at the moment of creation of each bubble universe the values of the physical constants can vary: the masses of the elementary particles, the strength of gravity, even the mathematical forms of the laws of physics including the number of dimensions of space! As a result, some of the bubbles do not expand, and some shrink back down to zero size. Our visible universe would be contained in one of the expanding bubbles.
All of the possible combinations - each of these bubble universes - is represented by a different point on the Landscape.
(This eternally inflating universe introduces a slight confusion of terminology: The entire structure containing all the bubbles is frequently referred to as the "universe", though my personal preference is to understand each bubble as a separate universe with the entire structure containing all the bubble universes being referred to as the "multiverse". This is in agreement with the semantic definition contained here. This is the terminology I use throughout this article.)
A Problem in the Multiverse?
Not everyone is a fan of the multiverse. One implication of the multiverse interpretation of the Anthropic Principle is that we no longer have to seek to find an elegant mathematical explanation of why the fundamental constants are set to their unique values. We only have to postulate the existence of a random process whose task is to which sets the constants to different random values in each universe - job done! This apparent side-stepping of difficult questions for which we currently have no answer is not welcomed by all. Brian Greene from "The Elegant Universe": "The Anthropic Principle has the capacity to lessen our insistence on explaining why our universe appears as it does". This echoes the concerns of Lee Smolin (discussed above).
Other critics of the multiverse have referred to the writings of the philosopher Karl Popper who believed that a theory had to be falsifiable or else it could not be considered a part of science. In his book of various essays, Are Universes Thicker Than Blackberries?, Martin Gardner delivers a scathing criticism of multiverse theories, quoting Popper: "Science does not proceed by induction - that is, finding confirming instances of a conjecture - but rather by falsifying bold, risky conjectures. Confirmation can be slow and is never certain. By contrast, a falsification can be sudden and definitive. Moreover, it lies at the heart of the scientific method." As the multiverse is stated to lie beyond the causal horizon of our universe, multiverse theories can never be proved wrong, and as such they should not be considered scientific.
I think it's fair to say that Paul Steinhardt is not a fan either: "I think it is far too early to be so desperate. This is a dangerous idea that I am simply unwilling to contemplate." (see here).
The Requirement for a Selection Process
One of the attractions of multiverse theories is that they appear to explain how the universe we inhabit takes the form it does. Why are the masses of the elementary particles set to their particular values? Why are there three spatial dimensions? It might appear that there has been some sort of selection process at some stage (not to be confused with a selection effect), selecting these values from a list of possible values. However, multiverse theories would appear to avoid any such requirement for a selection process: it is only in our particular universe that the physical constants are set to these values - in another universe they would all be set to different values. Thus multiverse theories appear to solve the selection problem.
However, there is something illusory about this neat solution. The selection problem has not been solved, in fact it has only been pushed back to a deeper, more fundamental layer.
At the core of why this problem arises is the principle that, while some features vary between each universe in a multiverse structure giving each universe its different characteristics, there must always be some invariant features that do not vary at all with each universe in a multiverse structure. These are the unchanging features which are necessary to generate each universe, including providing the underlying substrate of the entire multiverse structure. And it is these invariant features which now apparently require selection (rather than the features in the higher layer) - the selection process has just been pushed back to a deeper layer.
Paul Davies refers to this problem in his recently-published comprehensive study of the Anthropic Principle "The Goldilocks Enigma". He refers to the invariant features as the "universe-generating laws": "In the standard multiverse theory, the universe-generating laws are just accepted as given: they don't come out of the multiverse theory."
As an example, the theory of eternal inflation (described in a box above) is a wonderfully neat and imaginative multiverse solution which can give an impression of being an entirely self-contained theory as to why the universe takes the form is does: random quantum fluctuations cause inflation of "bubble universes", and those same random quantum fluctuations result in each bubble universe having different characteristics. It's all very neat, giving an impression of "having all the answers". However, the driving force behind eternal inflation - the constant feature in each bubble universe - is that random quantum fluctuation process governed by the Heisenberg Uncertainty Principle (the bubble universes being created by quantum nucleation). And that leads to the question: "Why does the quantum randomisation process take the form is does? Might it not have been a non-random process instead? What selection process sets the Planck constant in the Heisenberg relation?" Or, to put it another way, "How come the quantum?", asked John Wheeler. Eternal inflation also assumes the presence of the underlying spacetime substrate (possessing 10 or 11 dimensions) - another invariant feature - spreading through each universe. Eternal inflation gives an impression of answering our questions about the selection process in the universe, but that is illusory - the requirement for a selection process has just been pushed back to a deeper, more fundamental layer.
This principle was obviously realised by the New Scientist interviewer in this discussion with Leonard Susskind: "So even if you accept the multiverse and the idea that certain local physical laws are anthropically determined, you still need a unique mega-theory to describe the whole multiverse? Surely it just pushes the question back?" (see here). Paul Davies also refers to the problem in "The Goldilocks Enigma": "Like the proverbial bump in the carpet, the popular multiverse models merely shift the problem elsewhere - up a level from universe to multiverse."
It is always possible to consider a deeper level than the multiverse and say "Why is this particular multiverse structure selected (over other possible multiverse structures)?" Do we now have to consider multiverses in which the invariant features of the substrate (Paul Davies's "universe-generating laws") can vary, take other forms, or not exist at all?! If so, then this would point to there being some requirement for a multiverse selection process!
So maybe we have to take a further step back ...
And now ... the Ultiverse!
If we are now considering the invariant features of the multiverse substrate we are now posed with the question: "Why does the multiverse have its particular universe-generating laws and substrate structure? Why this multiverse rather than some other multiverse with a different structure?" To get round this new multiverse selection problem maybe we have to take a step backwards to a deeper layer and start considering a multiverse of multiverses, with the underlying structure being different in each multiverse.
But now we seem to be in an infinite regression. Do we have to consider multiverses of multiverses of multiverses of multiverses? There seems to be no way to avoid the selection process, it just keeps getting pushed further and further back.
So how do we avoid this selection problem? Maybe the answer is to keep stepping backward until we consider the most fundamental level of all, and then presume this is completely unconstrained: the "ultimate multiverse" (or ultiverse, to coin a phrase). Just say "Absolutely anything can and will happen". That would appear to be a way to avoid the selection problem: avoid making any selection at all! More ambitiously, it might also provide a tentative solution for a so-called "Theory of Everything".
This "anything goes" ultiverse structure was first proposed in 1986 by David Lewis in book The Plurality of Worlds. He suggested that every logically consistent universe actually existed (see modal realism).
Andrei Linde has also considered this unconstrained situation: "In some countries, everything that is not explicitly allowed, is forbidden. In some other countries (and in science), everything that is not explicitly forbidden, is allowed. We live in one of such countries, so why don't we use the freedom if it does not make us any harm?" (see here). (But, as we shall see later, it is precisely these constraints which are essential to (forms the essence of) physical reality).
John D. Barrow in his book Pi in the Sky had a similar idea when he considered that our universe could be imagined as a simulation running on computer hardware. As such, the type of computer hardware on which our "program" runs is irrelevant - we can be considered to be pure mathematical states, inhabitants of a mathematical "Platonic" realm (see the page on The Mathematical Universe for a fuller discussion of mathematical Platonism). According to John D. Barrow: "This approach has all sorts of interesting ramifications. It means that anything that can happen - anything that is a possible consistent statement in the language of mathematics - does happen in every possible sense of the word". This theme was continued by Max Tegmark: "All structures that exist mathematically exist also physically" (arXiv paper gr-qc/9704009).
These previous references to "logically consistent universes" (Lewis) and "consistent statements" (Barrow) are crucial: if this proposed ultiverse is to successfully exist in physical reality then it must be logically self-consistent (physical reality can have no inconsistencies, no contradictions or paradoxes - see the page on The Mathematical Universe for a discussion of logical consistency). But if the ultiverse is defined as the place where "absolutely anything can and will happen", will this necessarily give a logically consistent result? Surely not. In the page on The Mathematical Universe it was explained how physical reality could be considered as being built-up from a set of fundamental axioms: an axiomatic system. If the ultiverse is defined as the place where "absolutely anything can and will happen" then the resultant physical reality is being built-up from every possible axiom! Such a system could surely never be logically consistent: paradoxes would be rife. Any statement could be proved true (and false!). In our completely unconstrained ultiverse we find ourselves having to postulate the existence of absolutely anything and everything. As an example of the problem in physical terms, our axiomatic system would have to include the existence of an object which destroys the entire ultiverse structure! Self-referential paradoxes render the whole ultiverse structure logically inconsistent.
So it would appear the ultiverse is a bit of a non-starter. We're back to square one.
It would appear that in the transition from unconstrained mathematics to physical reality there must be some selection process which chooses only certain logically consistent systems (from the full range of mathematical possibilities). In fact, that selection process could be considered the very definition of physical reality!
For example, if you were creating a computer simulation of a universe you would want to avoid all possible inconsistencies (which would render your simulation impractical), so you would need to apply certain constraints. For example, in "The Need for Constraints" in the page on The Big Brother Universe a detailed example is presented of how you would need to prevent your "Sims" characters within your simulation from travelling backwards in time. This would be necessary to prevent "killing your own grandfather"-type paradoxes. In the example it was shown how such logical inconsistencies would manifest themselves as infinite loops in your computer simulation, crashing the entire simulation!
It is essential to apply constraints in order to produce a tenable, valid universe. And it is those constraints which define physical reality from the list of all unconstrained, inconsistent possibilities. To re-quote Andrei Linde: "In science, everything that is not explicitly forbidden, is allowed" (a view based on Murray Gell-Mann's Totalitarian Theorem: "In physics, anything that is not forbidden is compulsory."). Hence, it is clear that it is the constraints which play the decisive role in carving physical reality from the space of infinite possibilities.
But where do these constraints arise? What is responsible for this "carving" of physical reality from all possibilities?
From Ultiverse ... To Nulltiverse!
We have just discussed the requirement for some sort of "carving" operation to define a logically consistent physical reality from the space of all possibilities. But what is supposed to be responsible for this selection operation? Perhaps a more attractive solution is if we do not consider a physical reality "carved" out of everything, but rather consider it to be "built-up" out of nothing. Instead of a situation in which absolutely everything exists without need of a reason, we would now have a situation in which absolutely nothing would exist without a good reason for its existence. Following in a long line of painfully-contrived names, we might call such a structure a nulltiverse.
So we now need to seek an ex nihilo (creation out of nothing) solution. But what could possibly spring into existence in such an empty nulltiverse structure? Well, only those objects which could create themselves could come into existence (I find this more palatable than the previously-discussed unconstrained creation "for no good reason" principles of the ultiverse). Hence there would be no need of any selection or "carving" process to identify physical reality: objects select themselves, springing out of nothingness. But what sort of objects can create themselves? And, most importantly, could our very universe have created itself?
For an answer to these questions, see the sections on The Origin of the Universe and Causal Loop Theories on the Cosmic Universe page.
Comments
Here you can add your comments about this
article.
Your comments will appear instantly on this page (underneath this
form):
CHANCE AND RANDOMNESS
Another problem with “The Goldilocks Enigma” is Paul Davies naive
assumption that there is such a thing as “Chance”. Davies is not alone.
We all begin by thinking we know what chance is. If a coin is tossed we
believe that “chance” will determine whether it will fall heads or tails. But
that is not really the case. To say that something is a chance event is really
to say that all the causal factors of the event are not known to us. If we
construct a machine to provide an identical impetus and angle to the toss of a
coin it will fall predictably. Modern mass production frequently relies on this
mechanistic predictability.
We all know what Unicorns are, just as we
all know what chance is. But neither really exists. Both Unicorns and Chance are
just epistemological concepts, ideas that can exist in our minds. They are not
ontological realities, entities that really exist. Nicolai Hartmann has shown
that “nothing in the world exists by chance in the ontic sense. Everything
depends on conditions and occurs only where these are fulfilled. (“New Ways of
Ontology” 1953, 70)
Randomness is not synonymous with “chance”. Other
known planets are products of the same matter and forces as Earth, but they are
all different from planet Earth.
The processes that produced these
planets are random in the sense that the intersection of a range of physical
laws, in different particular material circumstances, gave rise to these
different outcomes. Similarly we can observe the random nature of the number
plates on vehicles on a highway, but we understand that different factual
circumstances in each case gave rise to this apparent randomness. Randomness is
not the result of “chance”.
My article "The Intelligent Design of the
Cosmos" criticizing Martin Rees' "Just Six Numbers", on the Philica.com site
apply also to The Goldilocks Enigma
- Dr.
A.B. Kelly, 27th December 2006
Thanks for an interesting post,
Anthony.
Referring to your statement "To say that something is a chance
event is really to say that all the causal factors of the event are not known to
us", it appears that quantum mechanics is fundamentally indeterminate, that it
is not a result of our incomplete knowledge of the causal factors (see the
section on the "Quantum Casino": http://www.ipod.org.uk/reality/reality_quantum_casino.asp
). I don't think you should assume there can be no such thing as fundamental
indeterminacy in nature.
I think you're being a bit harsh on The
Goldilocks Enigma. Paul Davies makes very few assumptions about "chance" or
anything else for that matter. That is the strength of the book.
"Rees
does not explain why a large number of universes would necessarily vary from one
another." This refers to the in-built assumptions of the multiverse theories
about the "invariant features" which I mentioned in my main article - Paul
Davies's "universe-generating" laws - and the unexplained space-time substrate
which you also refer to.
Neither Martin Rees's three possibilities or
your own paper appear to consider the possibility of a universe which generates
itself, setting its own parameters for life, something along the lines of John
Wheeler's "participatory universe". See my section on the "Tower of Turtles": http://www.ipod.org.uk/reality/reality_small_world.asp
For interested readers, here's a link to Dr. Kelly's paper: http://philica.com/display_article.php?article_id=50
- Andrew Thomas, 29th December
2006
Hi, I've studied the anthropic physics in some depth, so you might want to check-out my blog: http://evolutionarydesign.blogspot.com/ Predictions that the Goldilocks Enigma makes about the observed universe: http://evolutionarydesign.blogspot.com/2006/09/goldilocks-enigma.html There is a form of pre-conceived prejudice, or "anticentrist dogma" (as Brandon Carter put it), that diverts scientists reactionarily away from any serious investigation into the first most apparent implications of the evidence that points toward a strong anthropic principle. Their first and only tendency is to find rationale to "explain away" evidence, rather than to give equal time to it, and that isn't science, it's politics. - island, 1st February 2007
"Resolving the Goldilocks Enigma" is on the Philica.com website, paper No. 87: http://www.philica.com/display_article.php?article_id=87 - Dr. A. B. Kelly, 13th March 2007
Andrew Thomas
Thank you for
your comments. Sorry I was so long getting back to you. I took account of your
comments in “Resolving the Goldilocks Enigma”.
You say:
”Referring
to your statement "To say that something is a chance event is really to say that
all the causal factors of the event are not known to us", the Copenhagen
Interpretation says that quantum mechanics is fundamentally indeterminate, that
it is not a result of our incomplete knowledge of the causal factors. I don't
think you should assume there can be no such thing as fundamental indeterminacy
in nature.”
We live in a stratified world, and each stratum e.g.
(sub-atomic), (matter), (life), has its own categories. To apply concepts
appropriate in one category to a different category of reality is what
philosophers call a category mistake. Indeterminacy at the sub-atomic level
cannot be applied in the world of matter where the Principle of sufficient
reason applies, where everything is, and is as it is, for a reason.
For a
universe to generate itself it would have to exist before it existed. Perhaps it
is self-existent. If so it is God. Self-existence is the attribute of divinity.
You quote Steven Weinberg saying: "I have to admit that, even when
physicists will have gone as far as they can go, when we have a final theory, we
will not have a completely satisfying picture of the world, because we will
still be left with the question 'why?' Why this theory, rather than some other
theory?"
I address the Why question. My resolution makes sense of the
evidence.
My papers are on my Web page at http://www.philosophy.27south.com/
any comments welcome.
Tony Kelly
- Tony Kelly, 3rd April 2007
Reading these fascinating studies and the posts, there seems to be one point that you touched on - in Veiled Reality - that may be the crux of the whole matter. You spoke of the underlying reality which we are unable to observe. Also, entanglement seems to contradict observed reality. Perhaps we are trapped in a "Veiled Reality" ourselves, and its the real world that we can't see. It seems that there is simple evidence of this, and that is, that we are trapped by the arrow of time. We cannot observe anything except what obeys the arrow of time. That this is a limited view of reality is evidenced by the behaviour of entangled particles. This takes us back to what was presented as long ago as St-Augustine, who said that it is pointless to consider what happened before Creation, because time, as a measure of before and after, was created along with everything else. If time is a dimension of the space-time fabric, then the question of "What caused the Big Bang?" is a meaningless question. At the same time, all the other universes disappear, because they are unnecessary - if there was no time - no before & after - before (or rather "outside") the moment the space-time continuum was created, then there was nowhere for them to be. Hard for us finite beings to comprehend, but not so hard if you view time as following the laws of a mathematical dimension. It raises many questions, such as causation, probability, etc., all of which are constrained by the arrow of time. - John, 4th June 2007