Read play

To get an Idea of what we want to look at in the play, we started looking at the play by using Wordhoard. While doing so we filtered the most (and least) used words in Hamlet.

The practical aspect of Wordhoard is that it not only shows the most/least used words but also if the words are under/overused compared to other Shakespearian texts. Words in black ore overused whereas words marked in grey seem to be used less in Hamlet than in other plays.

[What we found interesting here are 2 observations:]{.underline}

  1. The lack of pronounces used in the play. Compared to other Shakespearian plays, the pronouns “I, she, thou, thy” seem to play a smaller role in Hamlet.

  2. “Madness” - as Wordhoard takes every word into consideration when doing its statistics, we have to consider ignoring the Names of the actors in the play as they are naturally always mentioned when speaking. Following this, we can see that one of the most used words in the play is “Madness”. This gave us a first indicator for something to dip deeper into. The meaning of Madness and the absence of pronouns in the play.

    Right after this we changed the modifications of Wordhoard a bit and extended the search to “Multiword Units”. This means Wordhoard showed us word combinations that appear most/least in the play. The number of words we used as an indicator was 2 to 5.

Paradoxical, this graph shows that one of the most used Multiword Units in the Play includes the Word “I” in the context of “I will” , which is one of the fewer used words compared to other Shakespearian plays.

We noticed that there is a lot of “Order” in the found data.

“My Lord”

“It be”

“I Will”

“There be”

All these words hint at contexts that are clearly defined. As “My Lord” is the most used Unit in the play, we see that the King (Claudius, murderer of Hamlets father) must be referred to in person a lot. It is also quite expected from a King to talk in clear and defined Orders as a King that talks vaguely might show weakness. Following this, we looked back at the plot: as Hamlet is a revenge play and the character wants to revenge his death father by any means, we wondered if there might be a connection with the data collected.

On one hand, we have the lack of Personal pronouns, such as

**” “ **

and the presence of Certainties and orders such as

**“It will, there be” **

or the presence of an authority known as

“My Lord”. On the other hand, we have the presence of the word “madness”.

In this context we wondered how these could relate to each other.

The first thing we observed is that the overuse of ” My Lord “ is not as specific as we thought. It is simply a form to address a male person in the times of Shakespeare, so almost basically everyone uses it to address his opponent while speaking in the play.

As we’ve seen when comparing Hamlet to the other tragedies, it can be noticed that the word 'I' is very rare. Using Wordhoard, we could see that while 'I' is underused, there are a variety of pronouns that are much more prominent.

Shakespeare's use of pronouns such as 'Thy', 'Thee' 'Thou' are important in knowing the status of the person being addressed. With these kinds of pronouns being rich throughout Hamlet we can gather a sense of the relationships between the characters in the play. Using the function Find words on Wordhoard, we were able to look up the specific words:

An interesting point that can be drawn from this evidence is how many times 'thou' is said in certain Acts. With thou being used in Hamlet to address an inferior such as King Claudius to servants and Queen Gertrude, we know it is an informal mode of address used for anyone who is of lower status than the speaker. ‘You', on the other hand, is a word used among nobles which we know is much more formal. Hamlet shows a switch between thou and you during his conversation with his ghost father. From the results above, Act I, Scene 5 thou only appears in the dialogue seven times. With this being a crucial point where Hamlet meets the ghost, it shows he does not yet view the ghost of a higher status than himself.

Conversely, going forward to the Act 3, Scene 5, Hamlet uses you to address his ghost father. In this scene, Hamlet does not seem that he sees ghost, rather he sees his father. Again, looking at the results above it indicates that thou is not spoken in this scene. This is useful evidence in showing that he now sees his father as superior and uses a formal mode of address. One interesting thought we concluded was the reason of the switching between thou and you is that he is convinced that the ghost is really his father whom he adores very much.

Looking at some of the results from Wordhoard the following show the comparing frequency of spelling (pn) in Hamlet compared to Shakespeare tragedies:

     Relative Use   Analysis Count   Log Likelihood   Analysis parts per 10,000 words   Reference parts per 10,000 words   ------ -------------- ---------------- ---------------- --------------------------------- ----------------------------------   Thee   \-             59               38.3             19.98                             41.96   Thy    \-             87               26.8\*\*\*\*     29.46                             50.21   Thou   \-             105              54.4\*\*\*\*     35.56                             69.65

    Relative use   Analysis Count   Log Likelihood   Analysis parts per 10,000 words   Reference parts per 10,000 words   ----- -------------- ---------------- ---------------- --------------------------------- ----------------------------------   I     \-             546              23.7\*\*\*\*     184.90                            228.78   You   \-             553              4.6\*            187.27                            169.58   He    \+             423              0.4              141.53                            146.06

Data explained:

  • The “Relative Use” column shows whether a word is more or less frequent in Hamlet compared to the rest of Shakespeare’s plays, just as the black/grey colour of the words show in the word cloud. Here we can see each of the pronouns is used less (minus in the table or grey in the word cloud).

  • The “Log Likelihood” column is much important. It shows the extent to which the frequency of a word in Hamlet in comparison to its frequency in the rest of Shakespeare’s plays. The number of asterisks shows the significance of the result, i.e. the more asterisks, the more difference between the frequency of the word in Hamlet compared to the rest of Shakespeare’s plays. Here, then, we can see that both “thy” and “thou” are used significantly less in Hamlet than in other Shakespearean plays.

  • The “Analysis parts per 10,000 words” column again shows the frequency of a word in Hamlet alone. For example, “thee” appears 38.3 times per 10,000 words.

  • The “Reference parts per 10,000 words” column performs a similar function, showing the frequency of the same word but this time across all of Shakespeare’s plays. E.g., “thou” appears 69.95 times per 10,000 words.


Using WordHoard allowed us to uncover the words most used within the play and forms them into a Word Cloud. From this, assumptions can be gathered regarding the key themes:

One of the most used words in Hamlet is ‘madness’ or words associated with this. The historical lemma of madness within Hamlet is high in comparison of all Shakespearean plays:

  • Madness: + 14.1**** - Analysis count: 68. Reference count: 22.

The comparison of many word forms also supports the idea that madness is a prominent theme by stating the word is overused in comparison to other Shakespearean texts (whereas ‘love’, for example, is underused). Another word that sticks out from the Word Cloud is ‘player’. This is interesting because it reflects the game of revenge which is central to Hamlet. The strategic use of madness is used to seek revenge.

There are several characters who could be categorized as mad, the most prominent two being Hamlet and Ophelia. The gender of madness and who speaks of madness throughout the play is of interest. Women are linked with madness only 603 times whereas male characters have 2974 linkages. It is easy to think of Hamlet when you think of madness within this play as he shows big signs of this. However, he originally feigns this madness in order to avenge his father. Though it is questionable to what extent this madness is real at the end of the play as it seems to encompass him. As a result, Ophelia’s madness worsens as she sees him deteriorate and she ultimately dies.

The “Find Word” function lets us recognize the key quotes surrounding the theme of madness:

Hamlet: ‘How strange or odd some’er I bear myself

(As I perchance hereafter shall think meet

To put an antic disposition on). (1.5.190-192)

- After the Ghost tells Hamlet that Claudius has murdered his father, Hamlet begins to plan his revenge. Here he describes his plan to pretend to be a madman, thus foreshadowing that he isn’t actually mad. However, Hamlet is already in a deep state of melancholy and unstableness. It is also of relevance to point out the Elizabethan idea of madness is different to modern day notions of mental illness.

Hamlet: ‘I am but mad north- north- west. When the wind is southerly. I know a hawk from a handsaw’. (2.2.352)

- Here we see Hamlet admitting he is not truly mad, but it is part of a plan. He knows madness from sanity and is acting like a fool to further his plan of gaining revenge for his murdered father.

Ophelia: ‘I would give you some violets, but they withered all when my father died.

They say a’ made good end’. (4.5.174-175)

- Throughout the play Ophelia shows more and more signs of a depressive state. She struggles to deal with her father’s death and continues to bring it up. Here she foreshadows her own death and hints that it is a ‘good end’ and possibly her solution to end the pain and suffering she is dealing with. The Queen states that Ophelia is ‘incapable of her own distress’ (4.7.177). The madness and lack of love from Hamlet probably does not help this.

How important is the notion of Madness, and what does Shakespeare try to tell us in the play?

1.) People affected by Madness


Gradual Development of his mental state from Played Madness to Mad Behaviour

-Graph of Moments where he acts “Insane” :

**[Data :]{.underline} **

Act 1:

Scene 5, 173-174 : I perchance hereafter shall think meet (To put an antic disposition on) : Hamlet says he’ll pretend to be crazy.

Act 2:

Scene 1, 84-95 : “He took me by the wrist and held me hard…At last, a little shaking of mine arm, And thrice his head thus waving up and down”

We see Hamlet mistreating Ophelia and acting weirdly, the first signs of his obsession and where his played madness seems to be not under control.

Scene 1, 83 : Polonius “Mad for thy love?” Assuming Love is the reason for Madness.

2.2.36 (My too much changed son) People assume reasons for his madness and even his mother sees he has “changed too much”.

**Act 3: **

Scene 4, 191-192 : Hamlet “ That I essentially am not in madness/ But mad in craft” He explains his mother that he is mad for real and that Claudius will try to expose him as a liar. *shows he is indeed not mad, as he explains himself)

Scene 4, 107 “Alas, he’s Mad” Gertrude when she witnesses Hamlet talking to himself (the ghost).

Scene 2, 370 : “ Let me be cruel, not unnatural..” Hamlet : shows he still has human instincts, he does not want to act unnatural or inhumane. His ability to reflect impose she is not mad (yet).

Scene 1, 163-165 : Claudius “Nor what he spoke, though it lacked form a little/ Was not like madness. There’s something in his soul/ o’er which his melancholy sits on brood..” Claudius does not think Hamlet is mad. He thinks Hamlet acts this way for a reason - anger to revenge his father.

Scene 1, 57 : “ To be, or not to be? That is the question/ Hamlet . The famous speech / monologue implies that he struggles to figure things out make decisions, it could be a strong indicator that he is losing substance, that he is not so sure of his plans and acting anymore.

Act 4:

Scene 4, 65 “Oh from this time forth, My thoughts be bloody, or be nothing worth!”

Hamlet decides to go after his goal. This determination shows his obsession and the trigger point for his madness: he wants to blank everything out, except for revenge.

Scene 1, 14 “His liberty is full of threats to all” Claudius.

Act 5:

“why, because he was mad” Gravedigger to Hamlet. Everyone thinks Hamlet is mad except for Hamlet himself, for thinks he acts like it.

“how come he mad?” Hamlet asking the Gravedigger.

Scene 2, 216-219” And you must needs have heard, how I am punished/With sore distraction. What I have done? That might your nature, honour, and exception/ Roughly awake, I here proclaim was madness” Hamlet says his actions result from madness.

Conclusion :

We can observe that it is never quite sure how much Hamlet is affected by a state of real madness and/or if he is still following his plan of faking it. What can be observed though is that the reason for his doubts and episodes of madness come from his obsession to revenge his father. It is the central point of his thinking and the main problem for him: he wants to kill Claudius by any means.

This shows how important revenge is in the play and how it affects everyone. Not only drives does it drive Hamlet mad, wo acts obsessed; it also turned Ophelia mad and led to her death as she could not handle the consequences of Hamlet’s actions (the death of her family members)

This lead us to the question: what did Shakespeare want to tell us about revenge?

Why is revenge portrayed negatively in the Play?

Why and how does Shakespeare portray revenge in such a negative way?

Here are a few ideas:

1) Leads to people going crazy :

As we have seen with Hamlet and Ophelia, revenge and its
consequences drive people into insanity. Revenge is bad as it
incites obsession.

Another example of this is the use of weapons in the play.

Shakespeare made wide use of poison. 4 out of the 8 people dying in
the play die of poisoning.

<img src="media/image5.png" />[^1]

The problems are:

1) To kill someone with poison is a cowards’ way of seeking revenge. As revenge in Shakespearian times should be a task to save the Honour of the murdered person, the way they should be seeking it is not by cowards’ methods.

It is an immoral way of killing. Morality however is often very undefined in the play. An example of this is the often used word “ good” , which is neutralized of any emotion in the play. One can say “good sir” which basically means ‘yes, sir’. It shows that things are not always clear in the play.

2) In the play, all that revenge does is leading to Misery and Death of affected and unaffected people.

  • 8 people die. All of these 8 people die related to Hamlets revenge plans.

3) Pretty much everyone dies, including people that should have been spared such as Gertrude (the ghost explicitly asked her to be spared) or Ophelia, who did no wrong. Shakespeare shows that revenge just leads to more misery, it is a vicious cycle.

4) Eventually it even leads to the Avenger’s death (Hamlet).

5) A lesson that can be learned is shown by the only survivor in the play, Horatio: he shows the Importance of forgiveness and rational thoughts of consequences. Horatio is the only survivor of the main characters in the play. We concluded that this might be because he is the only character acting within arguments of reason and forecast. Instead of getting obsessed over things, Horatio tries to stay logical and fix problems. He tries to warn Hamlet and solve problems instead of running into misery. From the early start of the play, he questions if what the ghost says is of truth and/or wise to follow. He does not give in to obsession such as Hamlet, who fails to think more critically and calm.

Hamlet :

Motive : Anger more than Honor : Determination Tunnel Vision Obsession Madness Tragedy (He Dies, everyone dies)

Leartes :

Honor Revenge Father Killed by Sword, Sister gone mad and drowned, Honourable reasons to take Revenge on Hamlet Death/Tragedy (Revenge has consequences even with good motives)

Horatio :

Rational Mind Thinking of Consequences Loyalty Acting right Survival (Lesson, Learned, revenge does not pay off.)

When looking at these traits of Horatio, who only has the best of his loyal friend Hamlet in mind, we tried to find any moments in the play in which Hamlet might have tried to solve his actual problem with Claudius instead of just raging to kill him.

To illustrate this, we used a program called “Sublime Text” which creates Character Network graphs. These show how much any characters interacted with each other:

Here we can observe that the amount of times Hamlet interacts with people is pretty evenly spread. There is no significant amount of time where he and Claudius actually interact and when looking deeper into the text and its content, one can see that, if they speak, it is never of much substance, nor does it constitute an attempt to talk about problems.

It shows that all this misery and death could have been prevented, if people would have just spoken more to each other, just like Horatio did.

Analysis through custom programs:

Additionally, we wrote 2 custom programs in Python in order to analyze various attributes of language. Both were designed with minimal set-up. So long as the text they were provided with was in a format where stage directions are marked with [], acts and scenes are marked with the word in block capitals followed by a line of = symbols, and character names are written in block capitals.

The first program:

The first of these programs focused on analyzing and comparing characters vocabulary to spot patterns.

The program created a table of all words featured in the play and how frequently they occurred, as well as a series of trees: one shared tree, and one for each character. Put simply, a tree is a data structure to show what a given piece of data can be followed by. For example, "well" might be followed by "done" or "met". The tree also stored how likely each branch was to be followed.

The tree data ultimately was not particularly useful for conducting analysis, although it could be put to use to "mimic" characters, with somewhat mixed results:

Although occasionally amusing, the trees were not viable for aiding analysis, thus the program was instead made to focus upon characters vocabularies. The first iteration of this looked at a sentence provided by the user, analyzed each word and decided who in the text provided (in this case Hamlet) was most likely to have said it. Words were not considered if they occurred too frequently, so as to minimize bias towards characters with more dialogue.

This worked best when it was fed quotes from other Shakespeare plays.

Polonius seemed to resonate with lady Macbeth. Generally, the tone and sentiment behind the quote and who it was matched to lined up quite well. Although it fared worse if fed modern texts:

Ultimately though, the trends it showed weren't especially useful, as you could simply analyze direct quotes from characters to gauge their general sentiment. We moved on instead to analyzing unique words. The program found all of each character's unique words, and then divided them by the total amount of words the character spoke, providing a percentage:

It is interesting how a character's unique word rate is more or less independent from how much they speak. For example, the king has a very low rate, despite having a lot of dialogue. Characters such as the Gravedigger and Doctor have higher numbers, although this is likely due to specific terminology combined with very limited presence in the play as a whole. Very generally, this seems to provide an estimated to how active a character is. Hamlet has the highest score because he interacts with many characters and takes a great deal of action, whereas the king mostly reacts and delegates to others, especially Polonius.

Although these results were somewhat informative, we had reached the limit of what language analysis alone could do. So, we started working on a second program.

The second program:

The second program was developed based on observations about existing programs used by Textlab; for the most part, they are all based upon the idea of taking raw text and using inbuilt functions to provide human-readable results and data. Whilst these are useful, this means subsequent programs must be built from the ground up. So, the second program was built with the idea of converting the raw text into a convenient format so that subsequent python programs could be written to understand and expand upon it.

The program is based on the following system:

  1. Split the play into acts

  2. Split each act into scenes

  3. Split each scene into lines

  4. Separate stage directions out

  5. Convert the lines into its composite parts: the contents, the speaker, and a unique ID.

This can then be written into a file and loaded into future programs. we wrote an example system into the program itself, which allows a user to input statements in a formal logical structure and returns statements that match. we then wrote a charting programing that took its output and displayed it. For example:


The taller bars show all the times that Hamlet and the King engage in conversation. It highlights that they mostly communicate in a few large conversations and implies the chaotic nature of the finale.



This is searching for an assorted grab-bag of violent words, so it's not surprising it mostly catches lines by the king and Hamlet; you can see when Polonius is killed when the Queen, who normally doesn't use violent words, uses some at the same time Hamlet does.

As discussed above though, the focus of this program is to provide an easy program with format for any given text, although the application of the graphing program does show that the data given to potential programmers does have ample potential for deep analysis.

Looking into Death in Hamlet

1) Compared to the amount of murder that occurs place in the tragedy AntConc showed that there is surprisingly few referrals to verbs relating to death i.e. murder kill or their derivatives. This lack of active words relating to death suggests that the text has a certain passivity towards death especially when compared to other tragedies. Othello for example has almost twice the hit rate of the word kill in comparison, despite having fewer deaths and murders.

2) Our Ubiquity results further proved the passive attitude Hamlet has as when we examined the scenes where murder and death occur in both Othello and Hamlet had a far lower rate of negative emotion.

3) Our WordHoard results showed that the word ‘player’ was of significance. The noun suggests a calculated and strategic attitude as identified by a group who had taken the class previously. However, we felt this further tied into our idea on how Hamlet portrays death with less emotion. We found using AntConc once more that hits on our concordance plot of ‘player’ generally line up with the hits in our concordance plot of the word ‘murder’. Which suggests that Hamlet conveys a more game-like calculated attitude towards death and murder rather than an emotional one.