Hi Peter, I am not sure about the role of resonating frequencies, but there has been some work on the relation between prosody and emotion recognition (particularly in the ASD literature). Here is an excellent article on prosody and the relationship with social power. You might find this useful. regards, Christian 

Uskul, A. K., Paulmann, S., & Weick, M. (2016). Social power and recognition of emotional prosody: High power is associated with lower recognition accuracy than low power. Emotion, 16(1), 11.

I think the effect that it has is that we can recognise the harmonic pattern and thus recognise the speaker. Someone who has heard my voice a few times will, upon hearing it again, pick up on the specific harmonic content and associate it with the memory of the previous times, even if I were speaking at different loudness or in a different mood.

Consider also that the emotive respones to major and minor chords may be learned responses, specific to certain cultures.

Whether or not certain historical figures had similar harmonic patterns, I'm not so sure, but they certainly used other aspects of speech to ellicit emotional responses in their audience, like prosody, as mentioned above. They also may have utilised up and coming broadcast technologies to allow their messages to be sent with higher fidelity than before, as radio and microphone technologies certainly improved in the 1930s.

The following articles may give you some cues, although they don't address this question directly.. You should familiarize yourself with the Journal of Speech, Language, and Hearing Research at http://jslhr.pubs.asha.org/

“Softened” voice quality in poetry reading and listener response.

doi: http://dx.doi.org/10.1075/ssol.5.1.03tsu

By Gafni, Chen; Tsur, Reuven

Scientific Study of Literature, Vol 5(1), 2015, 49-82.

In the performance of poetry, some performers occasionally use a certain ‘softened’ voice quality, deviating from that of the context, for emotive expression. We explored how listeners perceive such instances through a combined methodology of eliciting open-ended descriptions and ratings of pre-defined terms on ordinal scales. The collected responses confirm our intuitions that listeners are sensitive to this voice quality, which gives rise to impressions, such as affection, yearning, and contemplation. The chosen methodology allowed us, through variations between verbal and numerical responses, to observe the interaction of this voice quality with other factors, such as semantic content and other salient voice qualities. Finally, the responses highlight the distinction between two delivery styles used by performers—an impassive, grave style, typical of poetry recitals, on the one hand, and a more emotional “theatrical” style, on the other hand. (PsycINFO Database Record (c) 2016 APA, all rights reserved)

The effect of voice quality on ad efficacy.

doi: http://dx.doi.org/10.1002/mar.20712

By Wiener, Hillary J. D.; Chartrand, Tanya L.

Psychology & Marketing, Vol 31(7), Jul 2014, 509-517.

Extensive research has examined how to make broadcast advertising as effective as possible. However, one area of advertising that has received little research is the effect of the announcer or spokesperson's voice on consumers’ reactions to the ad or the product being advertised. In this paper, we examine the effect of the announcer's voice quality, or an overall impression of his or her voice, on consumers’ purchasing intentions for a new service. We find that female listeners are most likely to purchase this service when they hear it advertised by a man who has a creaky voice, but least likely to purchase it when they hear it advertised by a woman with a creaky voice. Other voice qualities produce more moderate results. Men, in contrast, are relatively insensitive to the effect of voice quality. © 2014 Wiley Periodicals, Inc. (PsycINFO Database Record (c) 2016 APA, all rights reserved)

Voice quality as a predictor of good rapport in the therapeutic alliance

doi: http://dx.doi.org/10.1037/e559712014-001

By Spinu, Laura E.; Sexton, James; Termini, Angela M.

2014 [American Psychological Association (APA)].

The importance of good rapport between a therapist and his/her client has been discussed in the literature extensively. Personal characteristics conducive to better rapport include the therapist's perceived honesty, warmth, trustworthiness and openness on the part of the client. Since, other than cues such as overall body language and gaze, voice has been shown to intrinsically encapsulate such characteristics, our study addresses the question of whether certain measurable physical properties of voice can be used to predict the development of a solid therapeutic alliance. We obtained permission from the clinic director at American University to access recordings of 40 therapist-client pairs. For each therapist, two 10-second portions of speech were extracted from each session for a total of 80 recordings. Acoustic analyses of voice quality (comprising overall pitch, pitch range, speech rate/tempo, fluency, glottal aspects such as breathiness and glottal fry, and certain formant components) were conducted. Next, the recordings were filtered using a low-pass Hamming filter, such that only aspects pertaining to overall intonation, but no linguistic content in the form of intelligible words, were preserved. The filtered utterances were included in a perception study. Thirty clinicians-in-training were asked to listen to each recording and rate it as either pleasant or unpleasant (forced choice) and also on a Likert scale based on several criteria (e.g. honesty, warmth, friendliness, trustworthiness, openness). We investigated statistically the extent to which these criteria correlate with the acoustic parameters we measured and identified a number of robust relationships. Secondary questions we address include (a) which of these characteristics are pervasive enough to be retrievable from low frequency content alone, (b) what specific voice characteristics are the strongest correlates to voice likability and (c) whether some of the criteria investigated are more important than others in making a decision on voice pleasantness. The findings can be used on their own to make predictions about therapist efficiency, incorporated to assessment tools (for instance, based on our results, we predict OQ45 improvement and retention), and further explored to establish voice training techniques for enhancing the relationship between client and therapist. (PsycEXTRA Database Record (c) 2014 APA, all rights reserved)

Parametric systems: The case of voice quality.

By Van Leeuwen, Theo

Jewitt, Carey (Ed). (2011). The Routledge handbook of multimodal analysis , (pp. 68-77). New York, NY, US: Routledge/Taylor & Francis Group, xxiv, 339 pp.

Culture and nature, meaning and materiality, have often been opposed to each other. In this chapter I try to argue for their inextricable connection. I offer an account of the semiotics of voice quality, of the kinds of meanings that may be conveyed by different voice qualities and the way these meanings are conveyed, and of the voices of iconic actors like Lauren Bacall, Marilyn Monroe and Marlon Brando, who have modelled what can be meant with voice quality. I will, on the one hand, try to show that there is nothing natural about the way meaning is assigned to voice quality. In different historical and cultural contexts, voice quality may play different roles and occupy a more or less prominent space in the semiotic landscape. On the other hand, I will also try to show that voice quality is ultimately grounded in something that is natural, the human body, the human vocal apparatus, an instrument built, not by human hand, but by nature. Culture and nature, meaning and materiality, have often been opposed to each other. I will also try to show that there are different kinds of semiotic systems. In systemic-functional accounts of multimodality, a single model, that of the 'system network' has generally been used. Here I will argue that, yes, there are, on the one hand, semiotic practices which can be described as system networks, or 'modes', as Kress and I called them in our book Multimodal Discourse, but on the other hand, practices that are better described as parametric systems, systems which articulate meaning directly in or with the material characteristics of the signifier, or 'media', as we called them in Multimodal Discourse. In short, I will, in this chapter, interweave the systematic-analytical and the cultural-historical aspect of multimodality, and I hope I will succeed in modelling how this can be done. (PsycINFO Database Record (c) 2016 APA, all rights reserved)