After reading through Sweeting's paper referenced in IEEE 1584-2018, "Arcing Faults in Electrical Equipment" (2011), I was left puzzled about the section addressing arc-flash impact on the human body. Specifically, he refers to tests conducted on pork legs showing that "[d]ue to the arc's negative v/i characteristic, any arc connection to the body self-extinguishes and is replaced with a glow column." Because of this effect, "resultant voltage across the body is likely to reduce the current across the heart to below ventricular-fibrillation levels." The assertion is that electric shock is not as impactful as burns from arc-flash (I think). My questions are:
(1) What is the negative v/i characteristic - voltage decreases as current increases?
(2) Why does this characteristic cause arcs to self extinguish?
(3) Why does voltage resulting from a glow column take electric shock out of the equation, leaving burns as the primary concern?

Thank you. Link to paper in IEEEXplore: https://ieeexplore.ieee.org/document/5625907

Dr Sweeting has been working on the arc flash topic with international papers from the 1990s. I have several of his papers including the pork leg paper that was also featured along with several other of his articles in various Australasian Power Technologies publications - Industrial Electrics and Transmission & Distribution. Dr Sweeting has also been active in presentations at various national and international engineering forums, several of which I have been fortunate to attend and in supporting industry with its understanding of power and high current matters.
My observations below are from my understanding based on my readings and previous discussions with Dr Sweeting, so my apologies if I have misinterpreted his meaning.
The reference to the negative characteristic relates to the impact of the arc voltage on the arcing current. Broadly he describes anode and cathode voltage drops of around 10V each which are material dependent (some papers suggest about 16V for copper), and a voltage drop in the arc column of around 10V/cm (some other papers suggest this might be 13V/cm) in ionising the gases. experimental plots of the arc voltage vs current show it to reduce slightly with increased arcing current. Conversely in an AC system, as the AC voltage approaches the zero crossing, the arcing current will be reducing and thus the arc voltage will be rising. This negative characteristic promotes the arc extinction till the voltage again rises sufficiently to restrike in any plasma that remains. If in lower voltage systems, the plasma dissipates before the voltage rises sufficiently (ie above the anode + cathode + 10V/cm gap), then restriking may not occur. (this of course is the principle that arcing horns in circuit breakers use in extending the gap and moving the plasma out of the arc chutes).
The pork leg experiments are described in the paper "Electrocutions and Arcs" (2008?) following observation that high voltage electric shocks do not always result in fatality but serious burns. Using a test voltage of 7.1kV, and high speed photography, the paper describes a series of 0.5s tests in which the shock current is seen to track across the surface of the pork leg causing an ionised air column in a ~0.1-3 milli-seconds. As a result of the 10V/cm gradient in this arc, it was hypothesised that the current cannot return inside the skin and thus fibrillation can be avoided. Further aspects of these tests included considering the effect of arc flash clothing for high voltage workers and whether such clothing design should promote moving the plasma outside of the clothing.

Thank you for the thoughtful and informative reply. I will take a look at the pork leg experiments described in that paper.