Diction: Palatalization

Polish and other Slavic languages have hard and soft consonants. Generally speaking, consonants can be softened, ergo palatalized, with the inclusion of “i” following a consonant or the diacritic kreska (e.g. ć) while some letters, such as “L” are always soft. Polish also contains what is called funkcjonalnie miękkie spółgłoskie (functionally-soft consonants). These consonants (c [ʦ], cz [t͡ʂ], dz [ʣ], dż [d͡ʐ], rz [ʐ], sz [ʂ], ż [ʐ]) were historically soft consonants but were later hardened. Now, they are viewed as a subgroup within hard consonants. In pronunciation, the articulation for these phonemes are in between those for their hard (lowest) and soft counterparts (highest) in terms of laminal proximity to the hard palate. Ronald Feldstein classifies the phonetic gradation in the following chart by using four classifications (t-, d-, s-, z-series) that are further separated into articulation types: dental, alveolar, or palatal. Other sources (Rocławski, Lulich, Hamann) regard the alveolar section as laminal retroflex consonants. In other words, these consonants (cz, dż, sz, ż/rz) should be articulated with the blade of the tongue rather than the body. While these sources simply cover spoken language and there is a miniscule difference acoustically between the alveolar (e.g. pronounced with body of the tongue) and retroflex pronunciations, the latter allows for an easier legato in singing. This post solely focuses on consonants in non-word final positions and without factors of assimilation.

Richard Miller describes legato as “the result of binding one sound to the next” achieved through the “continuous flow of vibrant vocal sound” (Miller, 122). This continuous flow stems from the uninterrupted changes in vocal timbre and air flow as the singer moves not just from note to note, but also phoneme to phoneme. If we step back from consonants and focus on the cardinal italianate vowels,[i] [ɛ] [a] [ɔ] [u], and how the air flow changes between the vowels one may feel the air is flowing at about a 45° angle from the hard palate on [i] and straight through the meeting of the hard and soft palates on [ɛ]. On [a] the air flow changes to the lowest area out of the vowel series, while maintaining the same direction as [ɛ]. [ɔ] is complicated because while being a back vowel in its nature, the air flow is perpendicular to the sound emanating from a singer’s mouth and [u] follows a similar path, but with a smaller focal point.

    One can apply a similar concept to the flow of air between hard to soft consonants. The dental consonants will follow the [ɔ]/[u] sensation, the retroflex of [ɛ]/[a], and palatal of [i]. (I hesitate to fully prescribe a location in which a person is supposed to ‘feel’ these sensations, but aim to highlight specific areas and points from which they may appear). To practice articulating these phonemes and maintaining an even airflow, I have made use of this following exercise. The exercise, entirely voiceless, is cumulative and when done successfully, the resonance space will not drop and the air will flow freely especially when incorporating vowels. First, begin with four iterations of [s], then [ɕ], and lastly [ʂ] in an easy 4/4. Next, four repetitions of the set [r] [p] [t] [k] followed by the first part. Lastly, a vowel resonance sweep from [u] to [i] over two beats. In this last exercise, the singer will glide from [u] to [i] while exhaling but without any vocal production (i.e. just form the vowel shapes, not the actual sounds). Repeat this twice (4 vowel sweeps total) followed by [r] [p] [t] [k] and the original sequence.

When examining descriptions of the phonemes [ʂ] and [ʃ], it becomes very apparent that these sounds are indeed quite similar in their pronunciation. Both phonemes are sibilant fricatives, meaning that air comes through a groove in the back of the tongue up to the articulation point, resulting in high-frequency turbulence. Moreover, both are pulmonic, central, oral consonants. The key difference is [ʃ] (the postalveolar fricative) is inherently partially palatalized while [ʂ] (retroflex fricative) is not palatalized, while remaining postalveolar. When forming the postalveolar fricative and the retroflex fricative, the tip of the tongue is the main articulator reaching behind the alveolar ridge, but differ in that the postalveolar involves the front of the tongue in a domed position while the retroflex phoneme is formed by a laminal tongue position (e.g. the tip of the tongue contact is flat). It is this domed position that causes the partial palatalization, which is not desired when pronouncing sz or its voiced counterpart, rz/ż. 

Conversely, palatalization is desired, and required, when pronouncing ś [ɕ] and ź [ʑ]. Once again, these sounds are sibilant fricatives and pulmonic, central, oral consonants. These phonemes are alveolo-palatal, meaning their articulation point is postalveolar, like [ʃ], and laminal, like [ʂ]. However, it is drastically palatalized. Therefore, the front of the tongue is raised towards the hard palate as if it is bowing. The tip of the tongue remains relatively loose and will be in the shape of whatever vowel follows the [i], which is actually the alternative spelling for ś and ź when followed by a vowel (się [ɕɛ] reflexive pronoun, ziemia [ʑɛṁa] “earth”).

When looking at an anatomical diagram, one can see how close the post-alveolar and retroflex articulation points are to each other. The main difference is the retroflex has the tongue almost folding back on itself while the post-alveolar makes the tongue rock forward. 

Acoustically, this difference in position will be darker for the retroflex because the airstream is suddenly interrupted by the tip of the tongue protruding upwards. The postalveolar fricative more or less has a similar obtrusion throughout the shape of the phoneme, which is brighter than the retroflex, but not as bright as the alveolo-palatal fricative. This phoneme is the brightest of all of the three, and is reflected in its high point of contact, which is just behind the retroflex. This acoustic difference is what makes words intelligible to native speakers. To become more accustomed to these acoustic differences, practice saying the following words: sztab [ʂtap] “staff”, siedem [ɕɛdɛm] “seven”, sam [sam] “by oneself”.

As stated before, the tip of the tongue is free to move because the articulation point in the alveolo-palatal fricative will maintain the brightness that distinguishes it from the aforementioned phonemes. Therefore the tip of the tongue will be in the shape of the upcoming vowel (i.g. siostra [ɕɔstra] “sister”, sąsiad [sɔ̃ɕat] “neighbor”, siódmy [ɕudmɨ] “seventh”). 

The same concept applies for the other functionally soft consonants (cz, dż) and palatalized consonants (ć, dź). The main difference is these consonants contain a stop before them (voiceless for cz and ć and voiced for dż and dź), which will be articulated in the same area as the fricative phonemes. It is important, however, to make sure that there is no overly pronounced shadow vowel when pronouncing words with dż (for example dżem [dʐɛm]. To perform this cluster, imagine that your tongue and the voiced sound is a continuous stream into the phoneme [ʐ]. One can also work backwards to isolate the individual phonemes and try to blend the cluster into a single gesture. 

In conclusion, there are three types of consonants in Polish: hard, soft (palatalized), and functionally/historically soft. The soft consonants are produced from a heightened tongue position that involves the body of the tongue being pressed against/close to the hard palate. Hard consonants, on the other hand, have a space below the hard palate and the functionally soft incorporates aspects of both. As singers begin to explore and become comfortable with these new sounds and sensations, one will find that these minute differences will help facilitate legato and more fluent diction because this style of pronunciation reflects the natural tendencies and phonetic flow of the Polish language. While practicing, singers should isolate the individual phonemes as in the exercise above and slowly work in the cardinal vowels, all the while building up to full words. In upcoming posts, this concept will be explored further in regards to devoicing, consonant clusters, and assimilation.