Context : HR 6819 was recently proposed to be a triple system consisting of an inner B-type giant + black hole ( BH ) binary with an orbital period of 40 d and an outer Be tertiary .
This interpretation is mainly based on two inferences : that the emission attributed to the outer Be star is stationary , and that the inner star , which is used as mass calibrator for the BH , is a B-type giant .

Aims : We re-investigate the properties of HR 6819 to search for a possibly simpler alternative explanation for HR 6819 , which does not invoke the presence of a triple system with a BH in the inner binary .

Methods : Based on an orbital analysis , the disentangling of the spectra of the two visible components , and the atmosphere analysis of the disentangled spectra , we investigate the configuration of the system and the nature of its components .

Results : Disentangling implies that the Be component is not a static tertiary , but rather a component of the binary in the 40-d orbit .
The inferred radial velocity amplitudes of K _ { 1 } = 60.4 \pm 1.0 km { s } ^ { -1 } for the B-type primary and K _ { 2 } = 4.0 \pm 0.8 km { s } ^ { -1 } for the Be-type secondary imply an extreme mass ratio of M _ { 2 } / M _ { 1 } = 15 \pm 3 .
We find that the B-type primary , which we estimate to contribute about 45 % of the optical flux , has an effective temperature of T _ { \mathrm { eff } } = 16 \pm 1 kK and a surface gravity of \log g = 2.8 \pm 0.2 [ cgs ] , while the Be secondary , which contributes about 55 % to the optical flux , has T _ { \mathrm { eff } } = 20 \pm 2 kK and \log g = 4.0 \pm 0.3 [ cgs ] .
We infer spectroscopic masses of 0.4 ^ { +0.3 } _ { -0.1 } M _ { \odot } and 6 ^ { +5 } _ { -3 } M _ { \odot } for the primary and secondary , which agree well with the dynamical masses for an inclination of i = 32 ^ { \circ } .
This indicates that the primary might be a stripped star rather than a B-type giant .
Evolutionary modelling suggests that a possible progenitor system would be a tight ( P _ { i } \approx 2 d ) B+B binary system that experienced conservative mass transfer .
While the observed nitrogen enrichment of the primary conforms with the predictions of the evolutionary models , we find no indications for the predicted He enrichment .

Conclusions : We suggest that HR 6819 is a binary system consisting of a stripped B-type primary and a rapidly-rotating Be star that formed from a previous mass-transfer event .
In the framework of this interpretation , HR 6819 does not contain a BH .
Interferometry can distinguish between these two scenarios by providing an independent measurement of the separation between the visible components .