Skip to main content
All Posts By

Jacob Thomas

RY/র‍্য Rendering Issues

The shaping of র‍্য (as in র‍্যাব) has often been a bug with Bangla Unicode fonts. According to the Unicode specification, the sequence should be:
rya sequence
Many Bangla Keyboard softwares use an incorrect sequence, especially using U+200C (ZWNJ) instead of U+200D (ZWJ); but this normally renders correctly anyhow.

InDesign & র‌্যা Shaping

In recent versions of InDesign (Oct 2022+), there has been a Unicode Bangla problem with র‍্য (as in র‍্যাব) displaying incorrectly as র্‌যা. This seems to be a bug in Lipika, Adobe’s Indic shaping engine. Fortunately, there’s a workaround. Adobe InDesign now supports the more advanced libre shaping engine HarfBuzz. If you activate HarfBuzz, this problem will be solved. Here’s how to activate it:

  1. Download this file: HarfbuzzOverride.js
  2. Place it in your InDesign /Scripts/Script Panel folder (Windows: C:\Program Files\Adobe\Adobe InDesign 2022\Scripts\Scripts Panel)
  3. Restart InDesign
  4. Go to Windows > Utilities > Scripts
  5. Select the HarfBuzzOverride.js file and double-click or ‘Run Script‘ from the window’s menu
  6. Now HarfBuzz should be activated and র‍্যা render correctly.

Thanks to John Hudson and Vinod Balakrishnan for helping me find these solutions.

Linotype Clones in Bangladesh

The last few months I’ve been trying to design a versatile and original Bengali text face (ie fonts for long texts), which has prompted a lot of research into earlier foundry types and questions about originality. Bengali text face design is a unique challenge, because virtually every Bangladeshi book and newspaper that’s been printed over the last thirty years has used a clone of Fiona Ross and Tim Holloway’s 1981 ABP Linotype Bengali design. I’m addressing two questions in this post:

  1. Is Mustafa Jabbar’s SutonnyMJ simply a clone of Ross & Holloway’s ABP Linotype Bengali?
  2. How does one develop an original and acceptable Bengali text face when an entire generation has seen nothing but one one text face?

In this post I’m going to focus in on the most widely-used typeface in Bangladesh, SutonnyMJ by Mustafa Jabbar of Bijoy Keyboard. This font is everywhere in Bangladesh – street signs, ephemera, books. Is it a clone of Ross & Holloway’s Linotype Bengali? To help you judge, I’ve placed the two types side-by-side here along with two pre-1981 text faces:

Besides being bolder than Linotype Bengali, there are little apparent differences as compared with previous typefaces pictured above. The only remarkable differences i have found so far are:

  • A much wider da দ ABP
  • Dropping the calligraphic connector stroke to the matra in a

Here’s another couple illustrations to help the reader compare SutonnyMJ with the original 1980’s Linotype Bengali designs:


clone-comparisons-03 As you can see, the letterforms are basically identical, with minor adaptions. Apparently Mustafa Jabbar had criticized Solaiman Karim a decade ago for copying his letterforms in the free font ‘SolaimanLipi’ (which are indeed crudely-made carbon copies!).

How different does a new text type have to be from its predecessors? How different are the following major Latin text faces?
clone-comparisons-latin-04I can quickly distinguish between SutonnyMJ and Linotype Bengali – does that mean Sutonny is an original design?

SutonnyMJ has a tremendous amount of issues, from overlapping glyphs to poor glyph shaping. Here’s a close-up snapshot of one glyph:


Plenty more could be said in criticism of this and other existing Linotype clones, but that’s beside the point.

How does one design an original, acceptable text face in an environment where any deviation from ABP Linotype is distracting to say the least? It’s a difficult tension – on the one hand, I don’t want to copy ABP Linotype, but significant deviations become so conspicuous.

I suppose it’s like how dominant Caslon’s type was in the eighteenth century – almost universally used in British printing.

Reph Positioning

One area in Bengali type where there’s no standard practice is in the positioning of the reph. The reph is an ‘r’ which precedes a consonant, and it is written as a diagonal slash above the consonant. However, some fonts center the reph over the letter, some align it with the connecting stem, others align it a certain distance from the letter’s left side bearing:

Reph Positioning_16

Reph Positioning_17Reph positioning above the character cha illustrates the differences most vividly. While most types position the reph on the left side of the cha, Adobe’s 2013 Bengali typeface aligns the reph with the left stem of the ‘cha’ (Fig 6).

From a legibility perspective, these are very significant differences, because to readers accustomed to seeing the reph to the right, Adobe Bengali’s alignment of the reph with the left-hand stem could seem to associate the reph with the preceding character and cause serious confusion.

It seems to me that the reph positioning around consonant clusters can affect legibility and meaning. So what is the best method for reph placement? To determine this, I looked at a wide variety of manuscripts to see what approach is used in traditional Bengali handwriting:

Reph Positioning_18While there is a lot of variety, it seems like the dominant approach is to align it with the right side-bearing of the previous letter, somewhat more centered for extra-wide letters. This is the case for virtually all Bengali fonts made in Bangladesh I have seen. This seems to also be the practice in historical forms of type, often due to technological constraints:

Reph Positioning_19historical-reph-samplesIn some historical type forms from a century ago, the reph is positioned radically to the right (“সম্পর্ক” above), which many present-day readers are still familiar with. This further corroborates our theory that alignment with the right side-bearing is preferable. [2019 note: I did find an example of more left-positioned rephs in a 1969 foundry type book by Indian Associate Publishing in Kolkata.]

However, following a contemporary strand of chirographic teaching, Adobe Bengali aligns the reph with the letter’s stem (whether left or right) but this approach seems problematic to me. Letters are viewed holistically, and the placement of the stem is visually insignificant, especially since a significant amount of letters do not even have a stem. Here’s a graphic that compares Adobe Bengali’s reph positioning with the standard practice in Bangladesh:

Reph Positioning_15 So my proposed solution is to set the reph generally aligned to the right side-bearing, but optically adjusted slight for wider letters and left-stemmed characters:

Reph Positioning_20Here’s some extra reph positioning examples thrown in from contemporary sign-painting and calligraphy in Bangladesh, though it really doesn’t have much bearing on this discussion:

reph-positioning-lettering[Added March 2019:] John’s comment below about a West Bengal publisher suggesting a left-aligned reph position brought to mind a similar revision with the 2017 Modern Bangla Dictionary from the Bangladesh Bangla Academy. The introduction states:

“কোনো ব্যঞ্জনের পূর্বে র যুক্ত হলে র-এর যুক্ত রূপকে রেফ বলা হয় এবং রেফ চিহ্ন ব্যঞ্জনের আরম্ভে দেওয়া সংগততর। এই যুক্তিতে বাংলা একাডেমি আধুনিক বাংলা অভিধান-এর বর্তমান সংস্করণে রেফ-এর অবস্থান একটু বাঁ দিকে সরিয়ে আনা হয়েছে।”
(my translation: “When a ‘ra’ is joined before another consonant its form is called reph, and placing it at the start of that consonant makes sense. Following this logic Bangla Academy’s current edition of the Modern Bangla Dictionary has moved the placement of the reph a bit to the left.”)

The actual implementation looks like this:

As my yellow highlights show, the font uses a fixed distance reph character regardless of letter width, which makes reph placement appear centered for wide characters (like জ) and extreme left for narrow characters (like ণ). It seems like an anchor attachment would have been more appropriate in this case. The BA’s newest dictionary also reverts to a lot of simplified conjuncts which often cause uneven color and overlaps.

Bengali Type Reading List

There’s not a lot written about Bengali type design, so I’d like to provide a reading list for anyone who is interested in learning more.

The Printed Bengali Character: Its Evolution (Fiona Ross)

Summary: This 200+ study by University of Reading professor Fiona Ross is a treasure for not only the study of Bengali typeface design but non-Latin typography in general. Together with Tim Hathaway, Fiona Ross were commissioned in the early 1980’s by Kolkata’s Anandabazar Patrika to redesign a Bengali typeface, and their design has subsequently become the quintessential Bengali typeface today, though usually under other pirated names. This volume traces the history of Bengali print typefaces, from calligraphic origins, through the very first initiatives by Charles Wilkins, through the prolific missionary presses at Serampore, documenting in detail the successive fonts and their relative merits and limitations. The study focuses especially on the way historical technologies’ constraints and misunderstandings formed the characteristics of each era’s fonts. The latter part of the book describes the process of designing Anandabazar Patrika’s font and the key role of the phonetic keyboard in the digitization of Bengali type. This volume has become a classic exemplary text, not only for Bengali but also for non-Latin typeface design background research in general.

Details: Ross, Fiona. The Printed Bengali Character: Its Evolution. London: RoutledgeCurzon, 1999. Book.

Availability: The only place I could find it was at the University of Minnesota Library

oloccoLinotype Bengali and the digital Bengali typefaces (Riccardo Olocco)

Picking up from 1982 where Ross’s Printed Bengali Character history left off, Olocco has researched the contemporary state of typeface design from 1982 until the present, highlighting the major typefaces. He begins with summarizing the earlierThere is a lengthy discussion of the issues in local typography, the lack of text typefaces, and the technological and institutional barriers to the emergence of quality Bengali type design in India. This is followed by a summary and critique of around ten Bengali typefaces designed in the last twenty years, most with connections to Western institutions. It is a great resource of documentation of recent type design initiatives – their strengths, shortcomings, and influences. Like most such studies, the research entirely omits the Bangladeshi scene where the majority of Bengali-speakers are. Nonetheless, this is an excellent source of background information on the major Indian and international Bengali typefaces of the last decade or two, and seems very thorough and relatively unbiased.

Details: Olocco, Riccardo. Linotype Bengali and the digital Bengali typefaces: With an enquiry into the current state of Bengali typesetting (Unpublished Masters Dissertation). University of Reading, 2014. Web. 7 Oct 2015.

Availability: full online download

milonAmar Bornomala o Shomoshamoyik Bangla Typography (Nazimuddaula Milon)

Written in Bengali, this 68-page MFA thesis describes Milon’s typeface design project for Axiata Telecom along with a number of other smaller typography projects he worked on. The central part is a broader discussion of the state of Bengali typography in Bangladesh, especially the technological and institutional challenges and the trends and projections for the future. He critiques the complacency of the local institutions (such as the Bangla Academy) for not taking more initiative in type development, and discusses constraints of prevalent typesetting technologies in Bangladesh. The paper concludes with projections for the future and a call for Bangladeshi designers to take more of an active role in the development of Bengali type. This thesis is a very good insight into the contemporary state of typography at Bangladesh’s leading graphic design program. Since I encountered this article a few months ago, I have met with the author and his professor at Dhaka University and discussed the article in more detail personally.

Details: Milon, Nazimuddaula. Amar Bornomala o Shomoshamoyik Bangla Typography (Unpublished Masters Dissertation). Dhaka University, 2010. Web. 11 Nov 2015.

Availability: Read online via here

non-LatinNon-Latin scripts: from hot-metal to digital type (Fiona Ross and Vaibhav Singh)

This book from the St Bride Library provides a detailed overview of the typographical issues of non-Latin scripts and typeface design. With articles by Paul Luna, Graham Shaw and Fiona Ross, the book gives extensive visual documentation from the St Bride Library collection of non-Latin typesetting. In the first section, Graham Shaw outlines the history of how various non-Latin scripts first encountered print and developed into typography. In the second section, Fiona Ross gives a visual chronicle of issues in non-Latin type design. The final section is especially relevant, an article by Fiona Ross on the key issues in type design for non-Latin scripts – the design brief, character repertoires, establishing dimensions, mark positioning and kerning, spacing, and harmonization with other scripts. The book finishes with a discussion of the importance of developing visual sensitivity and the value of calligraphy practice and research into manuscripts and early type design as a source for inspiration and guidance.

Details: Ross, Fiona and Singh, Vaibhav (eds) Non-Latin scripts: from hot-metal to digital type. London: St Bride Library, 2012.

Availability: I had to order it directly from St Bride’s Library in the UK and have it sent to the US.

Going Global: The Last Decade in Multi-Script Type Design (Gerry Leonidas)

From a faculty member of the world’s leading typeface design program this overview of current issues in non-Latin type design is a treasure of insight. It outlines the stages of global typeface design from the predominance of Latin forms and perspectives to the blossoming of research which has liberated these traditions. In the first stage (‘getting fundamentals right’), we learn about how designers and major institutions began to question the hegemony of Latin scripts and seek to develop non-Latin scripts. The second stage (‘linear families’) saw publishers coming to the forefront in commissioning non-Latin scripts, needing not just one font but families with a range of weights and widths. At stage 2 ½, minority scripts are being developed, non-Western type foundries are flourishing, and innovative typographic families of fonts are maturing. This article is a treasure for indicating from an authority in the field what the trends are in non-Latin typeface design.

Details: Leonidas, Gerry. “Going Global: The Last Decade in Multi-Script Type Design.” Typographica. October 15, 2013. Web. 7 Oct 2015.

Availability: Online via here.


Digital typeface design and font development for twenty-first century Bangla language processing (Fiona Ross)

The most recent publication from Fiona Ross, this fifteen-page article provides a summary of the key considerations that factor into successful contemporary Bengali typeface design. These include technological restrictions, the design brief, character set, character fitting, typeface dimensions and tools. A very helpful presentation of the design issues unique to Bengali typeface design, with much material that is not available in her other publications. Of special interest are the factors involved in considering character fitting, such as the significance of extremely high or low conjuncts in defining the overall height. Ross explains the different types of type for newspapers, books and screens, and how low-contrast and high-contrast fonts fulfill the criteria for each type application. Also the author provides her perspective on the key letters to begin a design with, the key factors and considerations of the design. The article is the opening part of a volume on Bengali language processing issues.

Details: Ross, Fiona. “Digital typeface design and font development for twenty-first century bangla language processing.” Technical Challenges and Design Issues in Bangla Language Processing (2013): 1-15.

Availability: You can see part of it via Google Books here, but the best way to get the full article is to buy that chapter direct from the publisher here ($30) and download the pdf.

Non-Latin typesetting in the digital age (Fiona Ross)

This article summarizes the unique issues that non-Latin typesetting faces, notably large character sets, complex positioning rules and therefore prohibitive design costs and technological constraints. She describes how in the nineteen century the trend for non-Latin scripts was to reduce the number of characters and simplify, and this led to an extreme dichotomy between chirography and print in India to the point where they were mutually unintelligible. The author charts the history of how these problems have been faced since the beginning of printing through the emergence of the digital age and now the contemporary scene. Since the emergence of digital and phototypesetting, the opportunities have expanded such that excellence in non-Latin type design are both desirable and realizable. Extensive examples are given from Bengali, Urdu, Thai, Devnagari and other scripts. This article shows some of the key benefits and opportunities that the digital age has opened up for Bengali typeface design.

Details: Ross, Fiona. “Non-Latin typesetting in the digital age.” Computers and Typography 2, Volume 2 (2002) Bristol: Intellect Books.

Availability: I could only find part of the article in google books here.

Image Making of the Letter Forms. Inspiration from Indian Image Making for Font Design (Bokila Prasad and Shilpa Ranadeb)

Expressing a need for new inspiration in Indic typeface design, the author proposes the Indian traditional proportions and grid systems of image/idol-making as an inspiration for typefaces. The first few pages describe how traditional idol-making in India uses a very specific grid system to manage and describe proportions. These insights are derived from the research of Alice Boner into medieval sculpture. The authors show the relevance this has for Indic scripts, in particular the use of an ‘angular grid’, meaning a target-shaped grid that is set at a defined angle. This is not exactly what is used in Indian sculpture, but a simplified version of it. This can help guide letterform formation in Bengali scripts. The article is a good expression of what current thought and practice is in the Indian University context regarding Indic typeface design inspiration, and as such reveals some areas of focus and exploration that haven’t been adequately pursued, such as contemporary calligraphy and lettering.

Details: Bokila, Prasad and Shilpa Ranadeb. “Image Making of the Letter Forms. Inspiration from Indian Image Making for Font Design” ICORD 11: Proceedings of the 3rd International Conference on Research into Design Engineering, Bangalore, India, 10 Jan 2011. Web. 10 Oct 2015.

Availability: Read or download in full from the website here.

Anatomy of Bengali Letterforms: A Semiotic Study (Subhajit Chandra, Bokil Prasad, and Udaya Kumar Dharmalingam)

Whereas Latin and Arabic scripts have a long-established vocabulary of typographic forms such as ‘bowl’ and ‘ascender’, such a vocabulary has not yet been developed for the Bengali alphabet to date. This short study attempts to define a grid and vocabulary of features and elements of the Bengali script using both anatomical and organic vocabulary. While this is a much-needed contribution to Bengali typeface design, I feel that their neglect of the Bangladeshi context weakens their solution. The majority of Bengalis live in Bangladesh, and it is beginning to take a leading role in Bengali typeface design, with a distinctly different style of typefaces in use, and yet this study claims the Bengali-speaking population there to be “15 million” rather than the actual 150 million. Furthermore, I think the choice of exclusively English terminology with no reference to Bengali terms hinders the widespread acceptance of these terms, especially since no discussion was made of pre-existing Bengali terminology (such as “উড়ানি urani” for the ascending swash curve).

Details: Chandra, Subhajit, Prasad Bokil, and Dharmalingam Udaya Kumar. “Anatomy of Bengali Letterforms: A Semiotic Study.” ICoRD’15 – Research into Design Conference. Jan 2015, IISc Bangalore. Web. 7 Oct 2015.

Availability: Read or download in full from the researchgate.netwebsite here.

Bengali Types and Their Founders (Katharine Smith Diehl)

More of historical interest, this article has largely been superceded by Fiona Ross’ research.

Details: Diehl, Katharine Smith. “Bengali Types and Their Founders”. The Journal of Asian Studies 27.2 (1968): 335–338. Web…

How the Giriśa Vidyāratna Press Acquired Its Fonts: A Supplement to the Work of Fiona G. E. Ross (Brian A. Hatcher and Fiona G. E. Ross)

A small supplement which explains some points that were missing from Fiona Ross’ earlier book.

Details: Hatcher, Brian A., and Fiona G. E. Ross. “How the Giriśa Vidyāratna Press Acquired Its Fonts: A Supplement to the Work of Fiona G. E. Ross”. Journal of the American Oriental Society 121.4 (2001): 637–639. Web…

Availability: JSTOR is where I read it.

A review of legibility studies and its implication to Indic scripts (Subhajit Chandra and D. Udaya Kumar)

This brief one-page visual study transfers Latin legibility research findings to the Bengali script, such as insights on stroke density, crowding, common area and terminals. It encourages the need for more study on Bengali legibility.

Details: Chandra, Subhajit and D. Udaya Kumar. “A review of legibility studies and its implication to Indic scripts.” Typography Day 2015 Conference. March 2015, IIT, Bombay. Web. 7 Oct 2015.

Availability: Download pdf from here.

Reading Bengali: A study of typeface readability and legibility for on-screen texts (Subhajit Chandra and D. Udaya Kumar)

An attempt to test legibility for Bengali scripts; I find the results hard to believe, since it ranks Vrinda as more legible than SolaimanLipi (a Linotype Bengali clone). Perhaps the participant quantity was not sufficient, I’m not sure.

Anatomy of Devanagari typefaces (Girish Dalvi)

Girish Dalvi is a professor of design at IIT Bombay, where he completed his PhD research in Devanagari Typography. He has designed a number of significant typefaces, including the Bengali typeface Star Bengali. Since unlike Latin scripts, Indic scripts do not have a standardized vocabulary for describing letterforms, this paper sets out to establish this vocabulary for the Devanagari script. The method used is to review and evaluate the vocabulary used to date by other previous authorities in the field, namely Bhagwat and Naik, M W Gokhale, and Mahendra Patel. The bulk of the article is describing the similarities and differences between the vocabulary of these authorities. The paper is relevant to Bengali typeface design because there is a need for a standardized vocabulary to describe Bengali letterforms, and often Devanagari nomenclature has been inappropriately used. This source helps to establish what terms are similar and which terms carry totally different meanings in these scripts (such as ‘matra’ and ‘shirorekha’).

Details: Dalvi, Girish. “Anatomy of Devanagari typefaces.” Des. Thoughts 1.1 (2009): 30-36.

Availability: Read or download the full pdf from the IIT website here.

Devanagari in Multi-Script Typography (Vaibhav Singh)

In this third millennium the frequency of multi-script texts both in Bangladesh and abroad necessitate the design of Bengali scripts which integrate stylistically with Latin scripts, yet this is a surprisingly elusive goal. ‘X-height’, headlines, overall color and stroke angle all inhibit integration of scripts. This dissertation focuses specifically on the combination of Devanagari scripts with Latin, but the underlying principles are very relevant for Bengali which is closely related to Devanagari. The author examines the role of the design brief, then technological constraints, and thirdly the history of how colonial governments and missionaries pioneered multi-script printing. Historical examples of multi-script families are critiqued and new insights drawn on key factors in multi-script families. It is very common for Bengali texts to contain words set in Latin type, and this resource is very helpful for guidance in regard to how to maximize compatibility and even color in such situations.

Details: Singh, Vaibhav. Devanagari in Multi-Script Typography (Unpublished Masters Dissertation). University of Reading, 2011. Web. 21 April 2016.

Availability: Read it on here.

Letterform Nomenclature

As I’ve been researching Bengali typeface design, I’ve been realizing that there’s a need for some standardized terminology or nomenclature for the elements of the Bengali script, both in Bengali and English. Since Fiona Ross’ The Printed Bengali Character is the seminal work in this field, I’ve used it as the basic standard. Chandra, Bokil and Kumar have attempted to clarify the nomenclature for Bengali type in a conference paper, but I don’t feel satisfied with it:

  • It seems to me they have dispensed with a lot of existing valid nomenclature that Ross and Polash Boron Pal and Chittranjan Bandopadhyay have already established in English and Bengali
  • They have introduced unusual terms like ‘delta’ instead of ‘triangle’, and ‘tail’ for an upper flourish; ‘wedge’ and ‘nose’
  • They have used Devanagari terms which carry a different meaning in Bengali, such as shirorekha for the headline. ‘Matra‘ is used for headline in Bengali; shirorekha means ascender line in Bengali. Matra means describes something else in Devanagari
  • They haven’t given much discussion of the existing Bengali language terminology

So I’ve spent some time researching terminology, browsing through The Printed Bengali Character by Ross, their study, Pal and Bandopadhyay’s terminology and colloquial terminology. Here’s what I would suggest as a work in progress:

Shrinking Conjunct Consonants

In Bengali (and other abugida scripts), multiply consonants are often clustered together such as follows:


The problem is, when combined consonants are stacked vertically they sometimes extend too far down, and therefore designers often will compress them to fit within the base height. However, this can be taken too far and make letterforms seem unnaturally compressed.

The best practice is clearly something between these two extremes of:

  1. either not compressing conjunct parts at all, or
  2. compressing them rigidly into the base height

To analyse best practices in this regard, I have compared below some of the major Bengali typefaces that are in use today in how they address these issues. The translucent red shapes are the original consonant shapes superimposed on the black consonant-conjuncts:

Conjunct Shrinking-01

It becomes clear that a lot of low-budget scripts designed in Bangladesh (SolaimanLipi and SutonnyMJ above) were formed by ASCII constraints, which meant not shrinking consonant conjuncts at all or shrinking them extremely (eg ত্ম). Vrinda falls into this same category. Adobe Bengali seems to emerge as the most nuanced and consistent in its treatment of conjunct resizing. Interestingly, a fount from the nineteenth century, Figgins Pica Bengali, seems rank higher than most all contemporary scripts in it’s care in resizing consonant conjuncts appropriately.

Bengali Italics?

In one sense, suggesting Bengali italics is as incongruously Eurocentric as speaking of Bengali ‘serif’, ‘sans-serif’ or ‘roman.’ It ignores the unique historical tradition of Bengali typography, and artificially slaps an alien historical tradition on top of it. Or does it? What I call ‘monolinear cultural evolutionism’ (the idea that all cultures will gradually evolve along the trajectory the West has) is clearly false and objectionable, and this mentality has led to a lot of silliness in typography. However, there’s also the overreaction of cultural purism or the ‘zoo’ mentality that attempts to dogmatically purge cultural traditions of all outside influences, and that doesn’t reflect the complex interrelations between cultural traditions.

I’d like to suggest that italics are an essential part of any quality Bengali typeface today, simply because they are so widely used and provide a layer of richness to the text. The introduction of punctuation into Bengali is a helpful analogy – early Bengali did not have spaces or punctuation, but the missionary printing presses introduced it and Vidyasagar popularized it, and now this ‘foreign influence’ has become an integral part of the Bengali language, albeit with slightly modified rules.

Although italics (তির্যক) have not traditionally been a part of Bengali typography, it is increasingly becoming mainstream according to English norms. Oblique forms are used by such convention-setters as the Prothom Alo (the most widely-read Bengali newspaper in the world), the Bangla Academy dictionaries (the authority in Bangladesh for spelling and grammar), and many other publications. However, to date these ‘italics’ are all simply oblique forms of the regular fount, thus harder for readers to differentiate from the regular text and ultimately hurting both legibility and beauty. The stylistic differentiation is important for visually distinguishing the italic text from the regular ‘roman’ type.

So I’d like to suggest an italic form for Bengali typefaces based on cursive Bengali script with the following common cursive characteristics:

  • a curvilinear headline/matra seems very appropriate for Bengali italics
  • a simpler rounded urani ascender which doesn’t curve upward at the terminal, and similarly simpler raphalas

Conveniently, last year I designed a script typeface that can be adapted as an italic version of the continuous text face I’m currently working on:


Bengali Sign-Painting

While in the West sign-painting is just a matter of nostalgia, in Bangladesh and India it is still produces the majority of signage.  Typeface design has a long-standing relationship with hand-lettered sign-painting, from Jan Tschichold’s heritage in sign painting to the popularity of retro display types today.  Can contemporary sign-painting be an inspiration for Bengali typeface design?  This article attempts to examine that question.

In my city of Rajshahi, there are well over a dozen sign-painting businesses, painting largely on fabric banners, metal signs and building walls.  Across this city I’d estimate that roughly half of all the text people encounter is hand-painted, showing how influential and pervasive it is.  While much of this hand-lettered type mimics existing computer fonts, there are regular interesting exceptions.  The most common embellishment distinctive to hand-lettered signs is the exaggerated forward curve at the bottom terminal of the verticals:

A few years ago a project in India called HandPaintedType attempted to create fonts based directly on sign-painting and sell them for $50 a font, but after three years it seems that it hasn’t developed very far.


Book Covers at the Dhaka Book Fair

As I’ve been researching Bengali calligraphy and display types, book covers can be a source for new inspiration, although the vast majority certainly just use Bengali Linotype font clones. So the day after একুশে ফেব্রুয়ারি (International Mother Language Day, Feb 21st) I braved the traffic with my camera to visit the Boi Mela, the country’s largest and best book fair.


Probably about fifty percent of book covers use the standard Linotype Bengali typeface, while another third use one or another of the locally available fonts, generally pretty poor quality. I didn’t see any typefaces from outside Bangladesh, such as Adobe Bengali, Nirmala, Noto Sans, or ITF’s Bengali typefaces.

There were a number of historical revival display typefaces, most with a feeling of ornament being arbitrarily tacked on and unrealistic calligraphic strokes.


Others tried to reproduce the feel of letterpress types from a previous era:



The hand-lettered titles mostly had either an extremely blocky feel to the point of being illegible or else excessive curls, which is something you see a lot in contemporary commercial graphics in Dhaka.



A lot of the calligraphic titles looked to be done with a square-tipped marker, and a few were quite well-proportioned and interesting.


Much of the calligraphy however has awkward letter-forms and an inconsistency of style and stroke:


The norm for Bangladeshi calligraphy has been to redraw the calligraphy as a vector shape, and then give it a slightly offset shadow. While this gives a decisive line, it loses the beauty of the hand-drawn calligraphy. Oftentimes the “vectorized” calligraphy is simply drawn as an weighted stroke (perhaps with Illustrator’s ‘calligraphy brush’), further reducing the expression of the calligraphy:

monolinear calligraphy

Many of the calligraphic titles were so poorly done that it would have been better to have used the standard Linotype Bengali text face:


As always, the late Qayyum Choudhury’s distinctive calligraphy outshines all the others in maturity and originality of style:


There were a few other unique and interesting finds out there: