Much recent work on tonal alignment, including its modeling, implies the existence of well-defined "targets" as well as "segmental" anchors to which the tones would be aligned. But researchers in this area are saddled with a paradox. In search for universal principles of alignment, they need to compare acoustic measurements for many languages, and actively search for effects that are specific to a single language, though a model which is itself specific to a single language is not desirable for a strong "universalist" view. However, most of the recent alignment work has been performed within the realm of acoustics, while perception and articulatory measures have been considerably neglected. After reviewing some acoustic studies, perception studies are presented which suggest that tonal targets might have a perceptual basis and might be guided by both psychoacoustic (i.e. universal) and language-specific constraints. Also, very recent articulatory data are presented which only partially support the strong hypothesis of universal articulatory constraints being the main driving force for alignment regularities.