Cu ce compari potrivirea?

Long-range linguistic comparison must explicitly model the probability of accidental resemblance; otherwise old cognacy and coincidence become hard to distinguish.

Evidence: Ungureanu frames the problem as distinguishing very old relationship from random coincidence after phonetic erosion and lexical replacement.

Inference: A comparison is not evidential merely because two short forms resemble one another.

methodological_proposal · methodological

Long-range comparison can, in principle, be made meaningful and falsifiable rather than remaining a collection of suggestive look-alikes.

Evidence: The paper explicitly seeks parameters under which long-range comparison remains possible and falsifiable.

Inference: The methodological ambition is to replace intuitive resemblance with testable constraints.

methodological_proposal · methodological

The published Reichenau 49/66 tally is highly inconsistent with an independent 50/50 Bernoulli null, conditional on that null and independence assumptions.

Evidence: 49/66=0.742; one-sided exact p=5.07386e-05 under p0=.5.

Inference: The descriptive skew merits explanation under the author's simplified model.

editorial_assessment · methodological

Treating Ungureanu's approximate 'we' result as exactly 15 matches in 50 trials gives an exact one-sided binomial tail of about 7.38×10^-5 under his p0=0.10 chance assumption.

Evidence: P(X≥15|n=50,p=.10)=7.3838686e-05.

Inference: The numerical excess is strong conditional on the author's null, but does not validate the null itself.

editorial_assessment · methodological

The later Altaic paper's literal false-positive arithmetic is internally inconsistent: 22 matches out of 600 is extremely unlikely under the paper's stated 1% independent false-positive null.

Evidence: Under Binomial(600,.01), expected=6 and P(X≥22)=3.174e-07.

Inference: The sentence that 22 barely exceeds the false-positive threshold cannot be justified by the written 1% null.

editorial_assessment · methodological

The 2010 article does not derive its 10% chance baseline for n-initial 'we' forms from a pronoun-specific control sample; the binomial result ceases to reach one-sided p<.05 if the true chance rate is about 19.5%.

Evidence: Threshold p0≈0.1949.

Inference: The baseline frequency is the key empirical quantity the original test needs to estimate.

editorial_assessment · methodological

Ungureanu's own earlier methodological calibration reportedly allowed 2–3% pairwise lexical coincidences; at roughly 2.5% the later paper's 22/600 count sits near one-sided p=.05, and at 3% it is compatible with chance.

Evidence: Review reports 2–3% pairwise coincidence rate; P(X≥22|p=.025)=0.0509; at p=.03, 0.198.

Inference: The later qualitative conclusion is more compatible with the author's earlier 2–3% calibration than with its stated 1% calibration.

editorial_assessment · methodological

The Reichenau paper does not establish that 0.5 is the correct independent-match baseline; at a baseline near 0.639 the observed 49/66 reaches one-sided p=.05.

Evidence: Exact sensitivity threshold p0≈0.639.

Inference: Control Romance languages are needed to estimate the null empirically.

editorial_assessment · methodological