Selection Without a Selector

A note on what this document is. It is an Institute paper, and it declines to use the Cosmic Cascade, as every Institute paper does. It does not argue that there are no Worldwrights. It argues that one observation offered in their support does not need them, and it says plainly what it cannot settle.

---

1. The observation

The record's comparison of its physics with the real universe's states that carbon-based life still dominates, "because Worldwrights optimise Concordants for chemical stability" (Physics Comparison, Matter and Biology). The observation is sound as far as this unit's data go. The Concordants it has data for are stable enough for chemistry, and in most of them chemistry has gone on to biology, by the several unrelated routes the survey has filed.

The question this paper asks is whether the observation is evidence for the explanation attached to it.

2. The selection effect

Every Concordant that has been measured was measured by someone standing in one. A survey party, an instrument and an archive all require a region stable enough for matter to hold together, for instruments to keep their calibration and for people to live long enough to read them. The sample of Concordants anyone has measured is therefore the sample of Concordants that could hold their measurers. Boundary zones are measured too, and their instability is filed at length, but they are measured from inside stable regions, by parties who came from one and go back to one.

A sample drawn that way would show stability whatever the population of Concordants was like. If most possible regions were unstable, the measured ones would still be stable, because the unstable ones hold no one to measure them. The physicist Robert Dicke made this argument about the age of the universe in 1961 CE (2161 UCSD), and Brandon Carter gave it its general form in 1974 CE (2174 UCSD): what we can observe is restricted by the conditions necessary for our presence as observers. Applied here, the stability of measured Concordants is a fact about the sample before it is a fact about the population, and a fact about the sample needs no one to have chosen it.

3. The measurement that would tell them apart

Selection and design make the same prediction about the existence of stability in the sample. They make different predictions about its margin.

Selection predicts stability that is enough and not much more. If measurers exist only where a region is stable enough for them, the measured regions should cluster just above that threshold, as a population thins towards the edge of what it can survive. Design for chemical stability, as the record states it, predicts a margin beyond what any measurer needs, because the stability was the point and not the precondition.

So the measurement is this: express each surveyed Concordant's stability as a margin above the threshold for sustained chemistry, in comparable units, and look at the distribution. Clustered near the threshold favours selection. Well clear of it favours design.

This measurement has not been made. The survey files each Concordant's coherence envelope, its stability and fertility conditions and its edge behaviour, but not as a margin above a common threshold, and this unit has not yet found a way to express them all in one unit without assuming the answer. The paper therefore proposes the measurement. It does not report a result, and anyone citing it as having found one has misread it.

4. What this paper cannot settle

Selection explains the sample, not the population. Even if every measured Concordant sits near the threshold, that would say why the regions we see are stable. It would not say why there are Concordants at all, or why the population includes any stable enough to hold anyone. This paper has nothing to say about that, and the record's own file on intent says the same in fewer words: insufficient basis for any finding.

Selection and design are not rivals. A sample shaped by selection may also have been shaped by a selector. Showing that one observation does not require design is not showing that there was none. The Institute's position on dragons has been criticised for treating a successful compression as a completed debunking, and the criticism is fair. This paper does not make that step. It shows that one observation offered for the design account does not need the design account, and it stops there.

The threshold itself is not known exactly. Where chemistry stops being sustainable is set by the survey's own filings at each site, and the filings do not agree on a common definition. The proposed measurement inherits that difficulty, and this paper does not solve it.

Some of the record's evidence for design is not this observation. The record's account of Worldwrights rests on more than chemical stability, and this paper addresses none of the rest.

— Dr. Oren Halsey, biosphere comparison unit, the Institute.

---

Works cited: R. H. Dicke, "Dirac's Cosmology and Mach's Principle", Nature 192 (1961 CE); B. Carter, "Large Number Coincidences and the Anthropic Principle in Cosmology", in M. S. Longair (ed.), Confrontation of Cosmological Theories with Observational Data (1974 CE).