My Visit to the Creation Forum at McLean Bible Church and a Review of "Hubble, Bubble, Big Bang in Trouble"
W.T. Bridgman, Ph.D.
On March 19, 2008, along with Dr. Jerry Jellison, I attended a meeting of the Creation Forum at McLean Bible Church in McLean, VA.
The primary topic of the meeting was showing the DVD by John Hartnett and Answers In Genesis, "Hubble, Bubble, Big Bang in Trouble" (2004). The group also mentioned Dr. Hartnett's more recent book, "Starlight, Time and the New Physics" (2007).
Dr. John Hartnett (http://internal.physics.uwa.edu.au/~john/) is a professor of Physics at the University of Western Australia. While he appears to have no professional training in astronomy (at least he demonstrated plenty of evidence of this in his DVD), he does have training in physics and does work in precision timing, particularly with crystal clocks that drive atomic fountain clocks.
On a historical note, in December of 2007, I encountered one of Dr. Hartnett's papers on redshift quantization posted on the Cornell Pre-print server (This does NOT mean the paper was peer-reviewed, though peer-reviewed papers can be posted there. The Cornell server merely serves as a common distribution system to scientists to disclose their research for comment by the community. Anyone who can get an existing endorser to vouch for them can get an account on the system). I sent him an e-mail pointing out the problems of power spectra in noisy data (having just completed a more detailed examination of the issue in by rebuttal of Donald Scott's book, "The Electric Sky"). Shortly after that, I discovered he was a Young-Earth creationist. To date, I have received no response from Dr. Hartnett.
Here I'll present some major points expressed in the DVD. I'll use black text to note Hartnett's points and boldface for my comments.
He does not mention the other important validation of the Big Bang: the observed abundances of the light elements (deuterium, helium, and lithium). These abundances match the Big Bang predictions to remarkable accuracy. This calculation is done starting with hydrogen and using the standard nuclear reaction physics in the expansion with only the baryon density as the free parameter (http://astro.berkeley.edu/~mwhite/darkmatter/bbn.html). The baryon density is found by matching the observed abundances (whose values vary across a factor of a BILLION) with the calculations.
The official names are the Hubble Deep Field (there is also the Hubble Deep Field South and the Hubble Ultra-Deep Field). The field of view is the same for most Hubble exposures on a given camera, so calling it the "Big Field" seems a little odd. Here are links to descriptions and more details about these observations:
He doesn't mention all the additional information we determine from spectra across many wavelengths. Redshift is just one of the many parameters we determine from spectra. Nor does he mention proper motion measurements or motions in binary systems which we use to discern dynamical characteristics of stars.
Cosmological redshifts apply outside the galaxy and technically in what is called the "Hubble Flow". The tradition of reporting redshifts as velocities grew out of the initial use of this information in determining velocities of objects within the galaxy, before we knew about extragalactic objects. The modern convention has moved away from reporting as velocity, which has ambiguities due to galactic motion and relativistic effects, and reporting strictly as 'z', which is the relative change in wavelength.
The Sombrero galaxy is actually classed as an unbarred spiral. The symmetric dark dust lane is a giveaway on that classification. This is an error that could be caught by anyone with an education in freshman-level astronomy. http://en.wikipedia.org/wiki/Sombrero_Galaxy
In the 1960s, Arp proposed that quasars were not cosmologically distant, but actually local objects ejected from the cores of active galactic nuclei (AGN, http://en.wikipedia.org/wiki/Active_galactic_nucleus). The standard explanation of the day was that quasars were actually the cores of distant AGN, viewed down the axis of the jet, at very high redshifts and distances. Although other observations increased support for the cosmologically distant interpretation of quasars (and became what is today called the 'unified model' of AGN), Arp continued to pursue his interpretation. Eventually the general astronomical community concluded that his claims were without foundation.
While M51 is in Arp's Atlas of Peculiar Galaxies (http://en.wikipedia.org/wiki/Atlas_of_Peculiar_Galaxies), it is not a discordant redshift pair. M51 (http://en.wikipedia.org/wiki/Messier_51) and its companion galaxy (NGC 5195, http://en.wikipedia.org/wiki/NGC_5195) actually are very close in redshift measurements and there is additional evidence of a past interaction between them.
In this case, there is no additional evidence of interaction between NGC 7603 and the 'nearby' quasars (http://adsabs.harvard.edu/abs/1986ApJ...302..245S, http://adsabs.harvard.edu/abs/2004A&A...421..407L). This one is particularly entertaining as the speckles in the image suggest the color table has been stretched to the point that noise is becoming visible. Like all extended and semi-transparent objects, if they are close together, there will be an enhancement between them as you receive photons from both objects.
This pair is popular since there is actual Hubble imagery (but before the first servicing mission), http://heritage.stsci.edu/2002/23/supplemental.html. Again, Hartnett shows an image that has apparently been colorized before their analysis, a process which loses image information. Their analysis may even have been generated from a color JPEG image, a format that is notorious for introducing artifacts due to the compression used. In this image also, the color table is stretched to the point that the speckling of noise appears. In the original Hubble image, there is no evidence of interaction between the two objects. I've examined the dataset myself (it is freely downloadable from the Hubble Data Archive) and evidence of a real, physical connection is poor at best.
Whatever it was, it was not co-registered with any optical image we could identify so real correspondences are impossible to see. To date, I have been unable to find any information on x-ray imaging observations of Mrk 205 by any mission. If someone knows of one, I'd appreciate some details.
Where did this astronomical 'principle' come from?? Did he just make it up? 'Closer' is a relative statement. Closer with respect to what? Usually, fuzzy means something is further away as the stars cannot be resolved. We do similar occultation observations of the Sun and other stars to see faint flows and dust around the object if they are lost in the glare. They often look 'fuzzy'. Does that mean the object is 'closer' than we think? This appears to the the observation he is referencing: http://hubblesite.org/newscenter/archive/releases/2003/03/image/b/. Here are some other high-resolution images of quasars showing 'fuzzy' structures around them: http://antwrp.gsfc.nasa.gov/apod/ap020309.html.
Of course, he doesn't explain the light-travel time issue. Arp's interpretation has an additional problem in that observations with more sensitive instruments reveal many background galaxies (or at least they appear 'smaller' in the images). Is Arp claiming these are all ejecta from the (larger?) foreground galaxy? Are 'baby galaxies' made up of 'baby stars'? Here's some examples of numerous 'small' galaxies around a 'large' galaxy:
Hartnett has posted some papers on the Cornell pre-print server on this topic. He makes the same types of errors computing power spectra as many others do, and as I did in doing time-series analysis, before I learned better (see the excerpt from my dissertation linked below). I plan to write more extensively on this topic in the future.
This ends my running commentary on the contents of the DVD presentation. Next I will discuss some additional issues raised by Dr. Hartnett along with their implications.
Like most creationists' talks, the key is in what they DON'T say. While explicit lying seems to be avoided, lies of omission and ambiguity are routine. Many of the erroneous statements could be dismissed as carelessness or the kinds of statements made by those ignorant of astronomy. Creationist cosmologists seem to rely heavily on what has become known as the "Ignorance Defense" in political and business circles: exploiting the fact that while lying can be a prosecutable offense, particularly in regards to obtaining money or power, being ignorant is not. This gives them impunity when it comes to accusing others of lying (for example, see some of Hartnett's statements about Hawking in "The heavens declare a different story!", http://www.creationontheweb.com/content/view/1657). With the level of mis-information and outright wrong statements ("Fuzzy means closer") in his presentation, should we judge Dr. Hartnett's errors as harshly as he judges Hawking? Is Dr. Hartnett's behavior consistent with the Golden Rule, one of the touchstones of Christian behavior?
It should be noted that Hartnett essentially accuses Hawking of lying in saying that there are no blueshifted galaxies when in fact there are. Hartnett does mention some blueshifted galaxies as noted above. However, I explored a few of the galaxy catalogs at CDS (http://vizier.u-strasbg.fr/cats/Cats.htx). A search of Rood (1980) with 3,981 galaxies finds 25 with blueshifts. The Millenium Galaxy Catalog (2003) has 69,594 objects (not necessarily all galaxies) and contains 40 objects with a blueshift and 'stellaricity' less than 0.5. The 2dFGRS catalog (2003) has 407,834 objects, contains 31 blueshifted objects with an 'eyeball classification' between 1 and 5, inclusive (for galaxy types). One needs to exercise care when exploring these catalogs because some contain objects that are not clearly galaxies. I've not yet figured out how to reliably make this selection for the Sloan Digital Sky Survey (SDSS) so that result is not included. Considering the miniscule number of blueshifted galaxies found in these LARGE surveys (and they are very small blueshift values compared to the large range of observed redshift values), I would say that Hawking made a reasonable approximate statement.
Many 'relativistic' creationists, such as Russ Humphreys or even Bob Gentry, try to integrate some of the time-altering effects of relativity into their pet cosmology. I have yet to find a research paper where they've fit their model with actual data to demonstrate that it works with a <10,000 year time scale. While Hartnett insinuates a <10,000 year time scale is possible in his cosmological model, I have yet to find any evidence, either in Hartnett's papers on the Cornell preprint server, or some of his papers available on creationist web sites, such as Answers In Genesis, where he can actually make a mathematically and physically consistent model fit the observations and time scale.
I can't emphasize this enough:
If Creation Scientists have successfully produced a relativistic model in a mathematically consistent form that gave a young universe (<10,000 years old) while keeping the Hubble relation intact, and didn't break significant other observational constraints, this would be a MAJOR development!
If creation scientists have done this, as they often insinuate, then where is the paper illustrating the solution? Remember that probably every graduate student in astrophysics has worked the Hubble relationship from the relativistic field equations in graduate school. It is NOT that difficult! If creationists haven't been shouting the young-universe solution parameters from the rooftops, it's probably because it doesn't exist.
I suspect Hartnett wants to avoid the mistake made by Russ Humphreys where providing actual details enabled real experts in general relativity, such as Don Page and Sam Conner (Conner is a fundamentalist Christian), were able to assemble a detailed list of Humphreys' errors. The detailed rebuttal version by Conner and Page, which analyzes the issues in intricate detail, is 200 pages long and is probably one of the best tutorials on general relativity I've ever read! This is understandable when you realize that creationists are not doing real science where their theories yield predictions that might be testable by more observations or even laboratory measurements. (See "The Cosmos in Your Pocket: How Cosmological Science Became Earth Technology", http://arxiv.org/abs/0710.0671). The goal of creation 'science' is not discovery, but to deceive the faithful.
Hartnett tries to claim that Halton Arp's discordant redshift observations falsify the Big Bang. What he doesn't explain is that that is only possible if they are not the result of chance alignments between near and far objects that only appear to be close together. Arp and co-horts use some very strange arguments for claiming these alignments are extremely low probability. The problem is that detailed examination of the volume of space included in these observations indicate the probability of chance alignments with distant background objects is actually much higher than they admit. Many of the 'discordant' pairs have redshifts that differ by a factor of 10 or more. For values of redshift and distance differing by a factor of ten, it is a simple geometric exercise to show that in a field-of-view around a given 'foreground' galaxy, the volume of space between the foreground and background galaxy is almost 1000 times larger than the space sampled from Earth to the foreground galaxy. In the Hubble Deep Field images, linked above, we see they are full of galaxies. Yet a single nearby galaxy can nearly fill the Hubble field-of-view. What is the probability of a 'chance' alignment under these conditions? Far larger than Arp and his supporters want to admit!
For more examples of how Arp's alignment argument is bogus, see Halton Arp's Discordant Redshifts.
In the post-presentation section of the forum, I raised the the point of how datasets with gaussian-distributed values can generate power spectra with strong peaks and that this is the error made by many proponents of redshift quantization. I had generated several graphics of quantized and non-quantized cosmological models which all displayed large peaks when processed through a power spectrum. I experimented with many of these issues in the temporal-domain in my Ph.D. dissertation. I've extracted the appendix of that document which contains an introduction to time-series analysis and time-series modeling: TimeSeriesAnalysis.pdf.
Some discussions moved into the literal interpretation of 'day' in Genesis. One person pointed out that the Hebrew use of 'day' in Genesis is a form that specifically means a 24-hour day. However, a short online search revealed this this interpretation has some issues ('The Hebrew Word "Yom" Used with a Number in Genesis 1' by Rodney Whitefield. http://www.godandscience.org/youngearth/yom_with_number.pdf). Another person did raise the issue of interpreting time with respect to God in regards to 2 Peter 3:8, "A day to God is as 1000 years...". Does a 6-day creation introduce another Biblical contradiction?
I did try ask my question of how would 14 billion years have been written in the language of the day, that is, could the writers of Genesis been divinely inspired but unable to write the numbers needed? Most of the number systems of the time and location that I know of don't represent anything above 1000, much less a million or billion. How would the ancient writers have written it? Another interesting possibility is presented in the very short science fiction story by Isaac Asimov, "How It Happened". The aforementioned speaker claimed the Hebrew 24-hour day interpretation totally invalidated the question, but I still regard the question as unanswered.
The main references for the redshift statistics are on my site under the Tiff & Arp pages at anomaly science.
- Wright, Edward L. Effect of intervening galaxies on quasar counts and colors. Astrophysical Journal, Part 1, vol. 311, p. 156-162. (1986). http://adsabs.harvard.edu/abs/1986ApJ...311..156W
- Newman, William I.; Haynes, Martha P.; Terzian, Yervant. Redshift data and statistical inference. The Astrophysical Journal, Part 1, vol. 431, no. 1, p. 147-155 (1994). http://adsabs.harvard.edu/abs/1994ApJ...431..147N
- Newman, William I.; Terzian, Yervant. Combinatorics and companion galaxies: Paradox lost. Astrophysical Journal, Part 1, vol. 441, no. 2, p. 505-506 (1995). http://adsabs.harvard.edu/abs/1995ApJ...441..505N
- Newman, W. I.; Terzian, Y.Power spectrum analysis and redshift data. Astrophys. Space Sci., 244, 127-142 (1996) http://adsabs.harvard.edu/abs/1996Ap%26SS.244..127N
References on digital signal processing and power spectra with references to power spectra and noise:
- J. S. Bendat and A. G. Piersol, "Random Data: Analysis and Measurement Procedures, 2nd Edition", Wiley-Interscience, 1986. [pp 282-286, 353-355]
- A. Papoulis, "Probability, Random Variables and Stochastic Processes", McGraw-Hill, 1991.
- S. Lawrence Marple, Jr. "Digital Spectral Analysis with Applications", Prentice-Hall, Englewood Cliffs, 1987 [pg 127-129].
Thanks to Jerry Jellison for proofreading and very helpful comments.
W.T. Bridgman, April 15, 2008