ggplot2 element_text() in R: Style Every Text Element

The element_text() function in ggplot2 styles every text element on a plot, titles, axis labels, legends, facet strips, and captions. You pass it inside theme() with arguments like size, face, color, family, angle, hjust, and vjust.

element_text(size = 14)                                # bigger text
element_text(face = "bold")                            # bold title
element_text(color = "navy")                           # color text
element_text(family = "serif")                         # change font family
element_text(angle = 45, hjust = 1)                    # rotate axis labels
element_text(margin = margin(t = 10))                  # add top margin
element_text(size = 12, face = "italic", color = "grey40") # combine styles

What element_text() does in ggplot2

element_text() builds a text-styling specification that ggplot2 plugs into any text slot. It does not render anything by itself. You always pass the result to theme() as the value of a text argument, like plot.title = element_text(size = 16).

Every text element in a ggplot2 plot has a name: plot.title, plot.subtitle, plot.caption, axis.title, axis.text, legend.title, legend.text, strip.text. Each one accepts an element_text() value. The helper makes styling reusable, you write one element_text(size = 14) and apply it everywhere.

Theme arguments inherit from broader categories. Setting text = element_text(family = "Arial") propagates the font to every other text element. More specific settings, like axis.text, override the broader ones.

Syntax and arguments

element_text(family, face, color, size, hjust, vjust, angle, lineheight, margin) covers every styling need. Each argument has a sensible default, so you only specify what you want to change.

The default theme renders all text in a moderate grey sans-serif. Every element you see, title, subtitle, axis labels, caption, legend, can be restyled with one element_text() per slot.

Seven styling patterns with element_text()

1. Bold, larger title

size is in points. face accepts "plain", "bold", "italic", or "bold.italic". Use these to set hierarchy across title, subtitle, and axis text.

2. Rotate x-axis labels

angle rotates around the label's center by default. Add hjust = 1 so the right edge aligns with the tick, otherwise labels overlap the axis line.

3. Center the title

hjust accepts 0 (left), 0.5 (center), 1 (right). The default for plot.title is 0 (left), which is why centering needs an explicit override.

4. Custom font family

Setting text = element_text(...) cascades to every text element. Individual settings still override the cascade. To use Google Fonts or system fonts, register them first via showtext::font_add_google() or extrafont::font_import().

5. Style the caption and subtitle

The default caption is right-aligned (hjust = 1). Set hjust = 0 to left-align it under the axis label.

6. Margin around text

margin(t, r, b, l) is in points. Use it when titles crowd the panel or axis labels collide with tick marks.

7. Style facet strip text

strip.text controls facet header text. Pair it with strip.background = element_rect(...) to restyle the surrounding box.

element_text() vs the other element_*() helpers

Each text slot accepts only element_text() or element_blank(). Other slots accept element_line() or element_rect(). Mixing them throws an error.

If you pass element_line() to plot.title, R errors with 'plot.title' must be an element_text object. Match the helper to the slot.

Common pitfalls

Pitfall 1: Forgetting theme(). element_text() returns a specification but renders nothing on its own. Always wrap it: + theme(plot.title = element_text(...)).

Pitfall 2: Wrong slot name. Use axis.title.x (lowercase, dotted). axis_title_x or AxisTitleX will fail silently in older ggplot2 versions and error in newer ones. Type ?theme for the canonical list.

Pitfall 3: Font not found. If family = "Helvetica" is not installed, R falls back to its default sans-serif WITHOUT warning. Register custom fonts first, or stick with "sans", "serif", "mono".

Pitfall 4: Confusing plot.title.position with hjust. hjust aligns within the title box; plot.title.position = "plot" (ggplot2 3.3+) aligns the title to the whole plot area, including the y-axis label space. Use the latter when you want a true left-edge title.

Try it yourself

Related ggplot2 functions

After mastering element_text(), explore these companions:

• theme(): the wrapper that accepts every element_*() spec
• element_line(), element_rect(), element_blank(): matching helpers for lines, rectangles, and hidden elements
• labs(): set the actual title, subtitle, axis names, and caption text
• margin(): build the margin spec for text padding and plot edges
• theme_minimal(), theme_classic(), theme_bw(): pre-built base themes to layer styles on
• showtext package: load Google Fonts and custom system fonts for use in family

For the canonical argument list, see the ggplot2 element_text reference.

FAQ

How do I change the font size of axis labels in ggplot2?

Pass element_text(size = N) to axis.text inside theme(). For example, theme(axis.text = element_text(size = 12)) sets both x and y axis tick labels to size 12. Target a single axis with axis.text.x or axis.text.y. Sizes are in points; 8 to 14 is the readable range for most plots.

Why is my custom font not showing in ggplot2?

R only sees fonts that are registered in its graphics device. Use extrafont::font_import() once to register system fonts, or showtext::font_add_google("Roboto") plus showtext_auto() for Google Fonts. Without registration, R silently falls back to its default sans-serif and your family setting is ignored.

What is the difference between hjust and vjust in element_text?

hjust controls horizontal alignment, 0 is left, 0.5 is center, 1 is right. vjust controls vertical alignment, 0 is bottom, 0.5 is middle, 1 is top. For rotated text, both axes rotate too, so hjust controls alignment along the new horizontal direction of the rotated label. Always preview after changing angle.

Can I use element_text() for plot annotations?

No. element_text() only styles theme slots (titles, axis labels, legends, facet strips). For text drawn inside the plot panel, use geom_text(), geom_label(), or annotate("text", ...), which accept size, fontface, color, and family arguments directly rather than through element_text().

How do I bold only part of a ggplot2 title?

Base ggplot2 does not support partial-bold strings. Use the ggtext package and replace element_text() with element_markdown() from ggtext. Then write your title as "Plot of mpg vs wt" with markdown bold syntax. The package parses the markdown and renders the bold portion correctly.